5.4.124-rc1 review

Signed-off-by Hui Wang <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 sound/pci/hda/patch_realtek.c |   23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

sound/pci/hda/patch_realtek.c +23 -0

@@ -2593,6 +2593,28 @@ static const struct hda_model_fixup alc8
 	{}
 };
 
+static const struct snd_hda_pin_quirk alc882_pin_fixup_tbl[] = {
+	SND_HDA_PIN_QUIRK(0x10ec1220, 0x1043, "ASUS", ALC1220_FIXUP_CLEVO_P950,
+		{0x14, 0x01014010},
+		{0x15, 0x01011012},
+		{0x16, 0x01016011},
+		{0x18, 0x01a19040},
+		{0x19, 0x02a19050},
+		{0x1a, 0x0181304f},
+		{0x1b, 0x0221401f},
+		{0x1e, 0x01456130}),
+	SND_HDA_PIN_QUIRK(0x10ec1220, 0x1462, "MS-7C35", ALC1220_FIXUP_CLEVO_P950,
+		{0x14, 0x01015010},
+		{0x15, 0x01011012},
+		{0x16, 0x01011011},
+		{0x18, 0x01a11040},
+		{0x19, 0x02a19050},
+		{0x1a, 0x0181104f},
+		{0x1b, 0x0221401f},
+		{0x1e, 0x01451130}),
+	{}
+};
+
 /*
  * BIOS auto configuration
  */

@@ -2634,6 +2656,7 @@ static int patch_alc882(struct hda_codec
 
 	snd_hda_pick_fixup(codec, alc882_fixup_models, alc882_fixup_tbl,
 		       alc882_fixups);
+	snd_hda_pick_pin_fixup(codec, alc882_pin_fixup_tbl, alc882_fixups, true);
 	snd_hda_apply_fixup(codec, HDA_FIXUP_ACT_PRE_PROBE);
 
 	alc_auto_parse_customize_define(codec);

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #2 lore raw

PATCH 2/177 ALSA: usb-audio: scarlett2: Fix device hang with ehci-pci

From: Geoffrey D. Bennett <[email protected]>

commit 764fa6e686e0107c0357a988d193de04cf047583 upstream.

Use usb_rcvctrlpipe() not usb_sndctrlpipe() for USB control input in the Scarlett Gen 2 mixer driver. This fixes the device hang during initialisation when used with the ehci-pci host driver.

Fixes 9e4d5c1be21f ("ALSA: usb-audio: Scarlett Gen 2 mixer interface")

Signed-off-by Geoffrey D. Bennett <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 sound/usb/mixer_scarlett_gen2.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

sound/usb/mixer_scarlett_gen2.c +1 -1

@@ -635,7 +635,7 @@ static int scarlett2_usb(
 	/* send a second message to get the response */
 
 	err = snd_usb_ctl_msg(mixer->chip->dev,
-			usb_sndctrlpipe(mixer->chip->dev, 0),
+			usb_rcvctrlpipe(mixer->chip->dev, 0),
 			SCARLETT2_USB_VENDOR_SPECIFIC_CMD_RESP,
 			USB_RECIP_INTERFACE | USB_TYPE_CLASS | USB_DIR_IN,
 			0,

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #3 lore raw

PATCH 3/177 ALSA: usb-audio: scarlett2: Improve driver startup messages

From: Geoffrey D. Bennett <[email protected]>

commit 265d1a90e4fb6d3264d8122fbd10760e5e733be6 upstream.

Add separate init function to call the existing controls_create function so a custom error can be displayed if initialisation fails.

Use info level instead of error for notifications.

Display the VID/PID so device_setup is targeted to the right device.

Display "enabled" message to easily confirm that the driver is loaded.

Signed-off-by Geoffrey D. Bennett <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 sound/usb/mixer_quirks.c        |    2 -
 sound/usb/mixer_scarlett_gen2.c |   79 +++++++++++++++++++++++++---------------
 sound/usb/mixer_scarlett_gen2.h |    2 -
 3 files changed, 52 insertions(+), 31 deletions(-)

sound/usb/mixer_quirks.c +1 -1

@@ -2268,7 +2268,7 @@ int snd_usb_mixer_apply_create_quirk(str
 	case USB_ID(0x1235, 0x8203): /* Focusrite Scarlett 6i6 2nd Gen */
 	case USB_ID(0x1235, 0x8204): /* Focusrite Scarlett 18i8 2nd Gen */
 	case USB_ID(0x1235, 0x8201): /* Focusrite Scarlett 18i20 2nd Gen */
-		err = snd_scarlett_gen2_controls_create(mixer);
+		err = snd_scarlett_gen2_init(mixer);
 		break;
 
 	case USB_ID(0x041e, 0x323b): /* Creative Sound Blaster E1 */

sound/usb/mixer_scarlett_gen2.c +50 ~~-29~~

91 lines not on this page · raw message

sound/usb/mixer_scarlett_gen2.h +1 -1

@@ -2,6 +2,6 @@
 #ifndef __USB_MIXER_SCARLETT_GEN2_H
 #define __USB_MIXER_SCARLETT_GEN2_H
 
-int snd_scarlett_gen2_controls_create(struct usb_mixer_interface *mixer);
+int snd_scarlett_gen2_init(struct usb_mixer_interface *mixer);
 
 #endif /* __USB_MIXER_SCARLETT_GEN2_H */

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #4 lore raw

PATCH 4/177 cifs: set server->cipher_type to AES-128-CCM for SMB3.0

From: Aurelien Aptel <[email protected]>

commit 6d2fcfe6b517fe7cbf2687adfb0a16cdcd5d9243 upstream.

SMB3.0 doesn't have encryption negotiate context but simply uses the SMB2_GLOBAL_CAP_ENCRYPTION flag.

When that flag is present in the neg response cifs.ko uses AES-128-CCM which is the only cipher available in this context.

cipher_type was set to the server cipher only when parsing encryption negotiate context (SMB3.1.1).

For SMB3.0 it was set to 0. This means cipher_type value can be 0 or 1 for AES-128-CCM.

Fix this by checking for SMB3.0 and encryption capability and setting cipher_type appropriately.

Signed-off-by Aurelien Aptel <[email protected]>

Signed-off-by Steve French <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/cifs/smb2pdu.c |    7 +++++++
 1 file changed, 7 insertions(+)

fs/cifs/smb2pdu.c +7 -0

@@ -979,6 +979,13 @@ SMB2_negotiate(const unsigned int xid, s
 	/* Internal types */
 	server->capabilities |= SMB2_NT_FIND | SMB2_LARGE_FILES;
 
+	/*
+	 * SMB3.0 supports only 1 cipher and doesn't have a encryption neg context
+	 * Set the cipher type manually.
+	 */
+	if (server->dialect == SMB30_PROT_ID && (server->capabilities & SMB2_GLOBAL_CAP_ENCRYPTION))
+		server->cipher_type = SMB2_ENCRYPTION_AES128_CCM;
+
 	security_blob = smb2_get_data_area_len(&blob_offset, &blob_length,
 					       (struct smb2_sync_hdr *)rsp);
 	/*

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #5 lore raw

PATCH 5/177 NFSv4: Fix a NULL pointer dereference in pnfs_mark_matching_lsegs_return()

From: Anna Schumaker <[email protected]>

commit a421d218603ffa822a0b8045055c03eae394a7eb upstream.

Commit de144ff4234f changes _pnfs_return_layout() to call pnfs_mark_matching_lsegs_return() passing NULL as the struct pnfs_layout_range argument. Unfortunately, pnfs_mark_matching_lsegs_return() doesn't check if we have a value here before dereferencing it, causing an oops.

I'm able to hit this crash consistently when running connectathon basic tests on NFS v4.1/v4.2 against Ontap.

Fixes de144ff4234f ("NFSv4: Don't discard segments marked for return in _pnfs_return_layout()")

Signed-off-by Anna Schumaker <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/nfs/pnfs.c |   15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

fs/nfs/pnfs.c +7 -8

@@ -1285,6 +1285,11 @@ _pnfs_return_layout(struct inode *ino)
 {
 	struct pnfs_layout_hdr *lo = NULL;
 	struct nfs_inode *nfsi = NFS_I(ino);
+	struct pnfs_layout_range range = {
+		.iomode		= IOMODE_ANY,
+		.offset		= 0,
+		.length		= NFS4_MAX_UINT64,
+	};
 	LIST_HEAD(tmp_list);
 	nfs4_stateid stateid;
 	int status = 0;

@@ -1311,16 +1316,10 @@ _pnfs_return_layout(struct inode *ino)
 	}
 	valid_layout = pnfs_layout_is_valid(lo);
 	pnfs_clear_layoutcommit(ino, &tmp_list);
-	pnfs_mark_matching_lsegs_return(lo, &tmp_list, NULL, 0);
+	pnfs_mark_matching_lsegs_return(lo, &tmp_list, &range, 0);
 
-	if (NFS_SERVER(ino)->pnfs_curr_ld->return_range) {
-		struct pnfs_layout_range range = {
-			.iomode		= IOMODE_ANY,
-			.offset		= 0,
-			.length		= NFS4_MAX_UINT64,
-		};
+	if (NFS_SERVER(ino)->pnfs_curr_ld->return_range)
 		NFS_SERVER(ino)->pnfs_curr_ld->return_range(lo, &range);
-	}
 
 	/* Don't send a LAYOUTRETURN if list was initially empty */
 	if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags) ||

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #6 lore raw

PATCH 6/177 iommu/vt-d: Fix sysfs leak in alloc_iommu()

From: Rolf Eike Beer <[email protected]>

commit 0ee74d5a48635c848c20f152d0d488bf84641304 upstream.

iommu_device_sysfs_add() is called before, so is has to be cleaned on subsequent errors.

Fixes 39ab9555c2411 ("iommu: Add sysfs bindings for struct iommu_device")

Cc [email protected] # 4.11.x

Signed-off-by Rolf Eike Beer <[email protected]>

Acked-by Lu Baolu <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Joerg Roedel <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/iommu/dmar.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

drivers/iommu/dmar.c +3 -1

@@ -1110,7 +1110,7 @@ static int alloc_iommu(struct dmar_drhd_
 
 		err = iommu_device_register(&iommu->iommu);
 		if (err)
-			goto err_unmap;
+			goto err_sysfs;
 	}
 
 	drhd->iommu = iommu;

@@ -1118,6 +1118,8 @@ static int alloc_iommu(struct dmar_drhd_
 
 	return 0;
 
+err_sysfs:
+	iommu_device_sysfs_remove(&iommu->iommu);
 err_unmap:
 	unmap_iommu(iommu);
 error_free_seq_id:

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #7 lore raw

PATCH 7/177 perf intel-pt: Fix sample instruction bytes

From: Adrian Hunter <[email protected]>

commit c954eb72b31a9dc56c99b450253ec5b121add320 upstream.

The decoder reports the current instruction if it was decoded. In some cases the current instruction is not decoded, in which case the instruction bytes length must be set to zero. Ensure that is always done.

Note perf script can anyway get the instruction bytes for any samples where they are not present.

Also note, that there is a redundant "ptq->insn_len = 0" statement which is not removed until a subsequent patch in order to make this patch apply cleanly to stable branches.

Example:

A machne that supports TSX is required. It will have flag "rtm". Kernel parameter tsx=on may be required.

 # for w in `cat /proc/cpuinfo | grep -m1 flags `;do echo $w | grep rtm ; done
 rtm

Test program:

#include <stdio.h>
#include <immintrin.h>

int main()

 {
        int x = 0;

        if (_xbegin() == _XBEGIN_STARTED) {
                x = 1;
                _xabort(1);
        } else {
                printf("x = %d\n", x);
        }
        return 0;

}

Compile with -mrtm i.e.

gcc -Wall -Wextra -mrtm xabort.c -o xabort

Record:

perf record -e intel_pt/cyc/u --filter 'filter main @ ./xabort' ./xabort

Before:

 # perf script --itrace=xe -F+flags,+insn,-period --xed --ns
          xabort  1478 [007] 92161.431348581:   transactions:   x                              400b81 main+0x14 (/root/xabort)          mov $0xffffffff, %eax
          xabort  1478 [007] 92161.431348624:   transactions:   tx abrt                        400b93 main+0x26 (/root/xabort)          mov $0xffffffff, %eax

After:

 # perf script --itrace=xe -F+flags,+insn,-period --xed --ns
          xabort  1478 [007] 92161.431348581:   transactions:   x                              400b81 main+0x14 (/root/xabort)          xbegin 0x6
          xabort  1478 [007] 92161.431348624:   transactions:   tx abrt                        400b93 main+0x26 (/root/xabort)          xabort $0x1

Fixes faaa87680b25d ("perf intel-pt/bts: Report instruction bytes and length in sample")

Cc Andi Kleen <[email protected]>

Cc Jiri Olsa <[email protected]>

Link http://lore.kernel.org/lkml/[email protected] open here

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/perf/util/intel-pt.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

tools/perf/util/intel-pt.c +4 -1

@@ -602,8 +602,10 @@ static int intel_pt_walk_next_insn(struc
 
 			*ip += intel_pt_insn->length;
 
-			if (to_ip && *ip == to_ip)
+			if (to_ip && *ip == to_ip) {
+				intel_pt_insn->length = 0;
 				goto out_no_cache;
+			}
 
 			if (*ip >= al.map->end)
 				break;

@@ -991,6 +993,7 @@ static void intel_pt_set_pid_tid_cpu(str
 
 static void intel_pt_sample_flags(struct intel_pt_queue *ptq)
 {
+	ptq->insn_len = 0;
 	if (ptq->state->flags & INTEL_PT_ABORT_TX) {
 		ptq->flags = PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_TX_ABORT;
 	} else if (ptq->state->flags & INTEL_PT_ASYNC) {

Greg Kroah-Hartman

author maintainer

2021-05-31 14:20 UTC #8 lore raw

PATCH 8/177 perf intel-pt: Fix transaction abort handling

From: Adrian Hunter <[email protected]>

commit cb7987837c31b217b28089bbc78922d5c9187869 upstream.

When adding support for power events, some handling of FUP packets was unified. That resulted in breaking reporting of TSX aborts, by not considering the associated TIP packet. Fix that.

Example:

A machine that supports TSX is required. It will have flag "rtm". Kernel parameter tsx=on may be required.

 # for w in `cat /proc/cpuinfo | grep -m1 flags `;do echo $w | grep rtm ; done
 rtm

Test program:

#include <stdio.h>
#include <immintrin.h>

int main()

 {
        int x = 0;

        if (_xbegin() == _XBEGIN_STARTED) {
                x = 1;
                _xabort(1);
        } else {
                printf("x = %d\n", x);
        }
        return 0;

}

Compile with -mrtm i.e.

gcc -Wall -Wextra -mrtm xabort.c -o xabort

Record:

perf record -e intel_pt/cyc/u --filter 'filter main @ ./xabort' ./xabort

Before:

 # perf script --itrace=be -F+flags,+addr,-period,-event --ns
          xabort  1478 [007] 92161.431348552:   tr strt                             0 [unknown] ([unknown]) =>           400b6d main+0x0 (/root/xabort)
          xabort  1478 [007] 92161.431348624:   jmp                            400b96 main+0x29 (/root/xabort) =>           400bae main+0x41 (/root/xabort)
          xabort  1478 [007] 92161.431348624:   return                         400bb4 main+0x47 (/root/xabort) =>           400b87 main+0x1a (/root/xabort)
          xabort  1478 [007] 92161.431348637:   jcc                            400b8a main+0x1d (/root/xabort) =>           400b98 main+0x2b (/root/xabort)
          xabort  1478 [007] 92161.431348644:   tr end  call                   400ba9 main+0x3c (/root/xabort) =>           40f690 printf+0x0 (/root/xabort)
          xabort  1478 [007] 92161.431360859:   tr strt                             0 [unknown] ([unknown]) =>           400bae main+0x41 (/root/xabort)
          xabort  1478 [007] 92161.431360882:   tr end  return                 400bb4 main+0x47 (/root/xabort) =>           401139 __libc_start_main+0x309 (/root/xabort)

After:

 # perf script --itrace=be -F+flags,+addr,-period,-event --ns
          xabort  1478 [007] 92161.431348552:   tr strt                             0 [unknown] ([unknown]) =>           400b6d main+0x0 (/root/xabort)
          xabort  1478 [007] 92161.431348624:   tx abrt                        400b93 main+0x26 (/root/xabort) =>           400b87 main+0x1a (/root/xabort)
          xabort  1478 [007] 92161.431348637:   jcc                            400b8a main+0x1d (/root/xabort) =>           400b98 main+0x2b (/root/xabort)
          xabort  1478 [007] 92161.431348644:   tr end  call                   400ba9 main+0x3c (/root/xabort) =>           40f690 printf+0x0 (/root/xabort)
          xabort  1478 [007] 92161.431360859:   tr strt                             0 [unknown] ([unknown]) =>           400bae main+0x41 (/root/xabort)
          xabort  1478 [007] 92161.431360882:   tr end  return                 400bb4 main+0x47 (/root/xabort) =>           401139 __libc_start_main+0x309 (/root/xabort)

Fixes a472e65fc490a ("perf intel-pt: Add decoder support for ptwrite and power event packets")

Cc Andi Kleen <[email protected]>

Cc Jiri Olsa <[email protected]>

Link http://lore.kernel.org/lkml/[email protected] open here

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/perf/util/intel-pt-decoder/intel-pt-decoder.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

tools/perf/util/intel-pt-decoder/intel-pt-decoder.c +5 -1

@@ -1090,6 +1090,8 @@ static bool intel_pt_fup_event(struct in
 		decoder->set_fup_tx_flags = false;
 		decoder->tx_flags = decoder->fup_tx_flags;
 		decoder->state.type = INTEL_PT_TRANSACTION;
+		if (decoder->fup_tx_flags & INTEL_PT_ABORT_TX)
+			decoder->state.type |= INTEL_PT_BRANCH;
 		decoder->state.from_ip = decoder->ip;
 		decoder->state.to_ip = 0;
 		decoder->state.flags = decoder->fup_tx_flags;

@@ -1164,8 +1166,10 @@ static int intel_pt_walk_fup(struct inte
 			return 0;
 		if (err == -EAGAIN ||
 		    intel_pt_fup_with_nlip(decoder, &intel_pt_insn, ip, err)) {
+			bool no_tip = decoder->pkt_state != INTEL_PT_STATE_FUP;
+
 			decoder->pkt_state = INTEL_PT_STATE_IN_SYNC;
-			if (intel_pt_fup_event(decoder))
+			if (intel_pt_fup_event(decoder) && no_tip)
 				return 0;
 			return -EAGAIN;
 		}

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #9 lore raw

PATCH 9/177 perf scripts python: exported-sql-viewer.py: Fix copy to clipboard from Top Calls by elapsed Time report

From: Adrian Hunter <[email protected]>

commit a6172059758ba1b496ae024cece7d5bdc8d017db upstream.

Provide missing argument to prevent following error when copying a selection to the clipboard:

Traceback (most recent call last):
File "tools/perf/scripts/python/exported-sql-viewer.py", line 4041, in <lambda>
menu.addAction(CreateAction("&Copy selection", "Copy to clipboard", lambda: CopyCellsToClipboardHdr(self.view), self.view))
File "tools/perf/scripts/python/exported-sql-viewer.py", line 4021, in CopyCellsToClipboardHdr
CopyCellsToClipboard(view, False, True)
File "tools/perf/scripts/python/exported-sql-viewer.py", line 4018, in CopyCellsToClipboard
view.CopyCellsToClipboard(view, as_csv, with_hdr)
File "tools/perf/scripts/python/exported-sql-viewer.py", line 3871, in CopyTableCellsToClipboard
val = model.headerData(col, Qt.Horizontal)
TypeError: headerData() missing 1 required positional argument: 'role'

Fixes 96c43b9a7ab3b ("perf scripts python: exported-sql-viewer.py: Add copy to clipboard")

Cc Jiri Olsa <[email protected]>

Link http://lore.kernel.org/lkml/[email protected] open here

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/perf/scripts/python/exported-sql-viewer.py |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

tools/perf/scripts/python/exported-sql-viewer.py +1 -1

@@ -2495,7 +2495,7 @@ def CopyTableCellsToClipboard(view, as_c
 	if with_hdr:
 		model = indexes[0].model()
 		for col in range(min_col, max_col + 1):
-			val = model.headerData(col, Qt.Horizontal)
+			val = model.headerData(col, Qt.Horizontal, Qt.DisplayRole)
 			if as_csv:
 				text += sep + ToCSValue(val)
 				sep = ","

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #10 lore raw

PATCH 10/177 perf scripts python: exported-sql-viewer.py: Fix Array TypeError

From: Adrian Hunter <[email protected]>

commit fd931b2e234a7cc451a7bbb1965d6ce623189158 upstream.

The 'Array' class is present in more than one python standard library. In some versions of Python 3, the following error occurs:

Traceback (most recent call last):
File "tools/perf/scripts/python/exported-sql-viewer.py", line 4702, in <lambda>
reports_menu.addAction(CreateAction(label, "Create a new window displaying branch events", lambda a=None,x=dbid: self.NewBranchView(x), self))
File "tools/perf/scripts/python/exported-sql-viewer.py", line 4727, in NewBranchView
BranchWindow(self.glb, event_id, ReportVars(), self)
File "tools/perf/scripts/python/exported-sql-viewer.py", line 3208, in __init__
self.model = LookupCreateModel(model_name, lambda: BranchModel(glb, event_id, report_vars.where_clause))
File "tools/perf/scripts/python/exported-sql-viewer.py", line 343, in LookupCreateModel
model = create_fn()
File "tools/perf/scripts/python/exported-sql-viewer.py", line 3208, in <lambda>
self.model = LookupCreateModel(model_name, lambda: BranchModel(glb, event_id, report_vars.where_clause))
File "tools/perf/scripts/python/exported-sql-viewer.py", line 3124, in __init__
self.fetcher = SQLFetcher(glb, sql, prep, self.AddSample)
File "tools/perf/scripts/python/exported-sql-viewer.py", line 2658, in __init__
self.buffer = Array(c_char, self.buffer_size, lock=False)
TypeError: abstract class

This apparently happens because Python can be inconsistent about which class of the name 'Array' gets imported. Fix by importing explicitly by name so that only the desired 'Array' gets imported.

Fixes 8392b74b575c3 ("perf scripts python: exported-sql-viewer.py: Add ability to display all the database tables")

Cc Jiri Olsa <[email protected]>

Link http://lore.kernel.org/lkml/[email protected] open here

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/perf/scripts/python/exported-sql-viewer.py |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

tools/perf/scripts/python/exported-sql-viewer.py +3 -2

@@ -122,8 +122,9 @@ if pyside_version_1:
 	from PySide.QtGui import *
 	from PySide.QtSql import *
 
-from decimal import *
-from ctypes import *
+from decimal import Decimal, ROUND_HALF_UP
+from ctypes import CDLL, Structure, create_string_buffer, addressof, sizeof, \
+		   c_void_p, c_bool, c_byte, c_char, c_int, c_uint, c_longlong, c_ulonglong
 from multiprocessing import Process, Array, Value, Event
 
 # xrange is range in Python3

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #11 lore raw

PATCH 11/177 perf scripts python: exported-sql-viewer.py: Fix warning display

From: Adrian Hunter <[email protected]>

commit f56299a9c998e0bfbd4ab07cafe9eb8444512448 upstream.

Deprecation warnings are useful only for the developer, not an end user. Display warnings only when requested using the python -W option. This stops the display of warnings like:

tools/perf/scripts/python/exported-sql-viewer.py:5102: DeprecationWarning:
an integer is required (got type PySide2.QtCore.Qt.AlignmentFlag). Implicit conversion to integers using __int__ is deprecated, and may be removed in a future version of Python.
err = app.exec_()

Since the warning can be fixed only in PySide2, we must wait for it to be finally fixed there.

Link http://lore.kernel.org/lkml/[email protected] open here

Cc Jiri Olsa <[email protected]>

Cc [email protected] # v5.3+

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/perf/scripts/python/exported-sql-viewer.py |    5 +++++
 1 file changed, 5 insertions(+)

tools/perf/scripts/python/exported-sql-viewer.py +5 -0

@@ -91,6 +91,11 @@
 from __future__ import print_function
 
 import sys
+# Only change warnings if the python -W option was not used
+if not sys.warnoptions:
+	import warnings
+	# PySide2 causes deprecation warnings, ignore them.
+	warnings.filterwarnings("ignore", category=DeprecationWarning)
 import argparse
 import weakref
 import threading

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #12 lore raw

PATCH 12/177 proc: Check /proc/$pid/attr/ writes against file opener

[1] https://www.kernel.org/doc/html/latest/security/credentials.html?highlight=confused#open-file-credentials

From: Kees Cook <[email protected]>

commit bfb819ea20ce8bbeeba17e1a6418bf8bda91fc28 upstream.

Fix another "confused deputy" weakness[1]. Writes to /proc/$pid/attr/ files need to check the opener credentials, since these fds do not transition state across execve(). Without this, it is possible to trick another process (which may have different credentials) to write to its own /proc/$pid/attr/ files, leading to unexpected and possibly exploitable behaviors.

Fixes 1da177e4c3f41 ("Linux-2.6.12-rc2")

Signed-off-by Kees Cook <[email protected]>

Signed-off-by Linus Torvalds <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/proc/base.c |    4 ++++
 1 file changed, 4 insertions(+)

fs/proc/base.c +4 -0

@@ -2556,6 +2556,10 @@ static ssize_t proc_pid_attr_write(struc
 	void *page;
 	int rv;
 
+	/* A task may only write when it was the opener. */
+	if (file->f_cred != current_real_cred())
+		return -EPERM;
+
 	rcu_read_lock();
 	task = pid_task(proc_pid(inode), PIDTYPE_PID);
 	if (!task) {

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #13 lore raw

PATCH 13/177 net: hso: fix control-request directions

From: Johan Hovold <[email protected]>

commit 1a6e9a9c68c1f183872e4bcc947382111c2e04eb upstream.

The direction of the pipe argument must match the request-type direction bit or control requests may fail depending on the host-controller-driver implementation.

Fix the tiocmset and rfkill requests which erroneously used usb_rcvctrlpipe().

Fixes 72dc1c096c70 ("HSO: add option hso driver")

Cc [email protected] # 2.6.27

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/usb/hso.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/net/usb/hso.c +2 -2

@@ -1689,7 +1689,7 @@ static int hso_serial_tiocmset(struct tt
 	spin_unlock_irqrestore(&serial->serial_lock, flags);
 
 	return usb_control_msg(serial->parent->usb,
-			       usb_rcvctrlpipe(serial->parent->usb, 0), 0x22,
+			       usb_sndctrlpipe(serial->parent->usb, 0), 0x22,
 			       0x21, val, if_num, NULL, 0,
 			       USB_CTRL_SET_TIMEOUT);
 }

@@ -2436,7 +2436,7 @@ static int hso_rfkill_set_block(void *da
 	if (hso_dev->usb_gone)
 		rv = 0;
 	else
-		rv = usb_control_msg(hso_dev->usb, usb_rcvctrlpipe(hso_dev->usb, 0),
+		rv = usb_control_msg(hso_dev->usb, usb_sndctrlpipe(hso_dev->usb, 0),
 				       enabled ? 0x82 : 0x81, 0x40, 0, 0, NULL, 0,
 				       USB_CTRL_SET_TIMEOUT);
 	mutex_unlock(&hso_dev->mutex);

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #14 lore raw

PATCH 14/177 mac80211: assure all fragments are encrypted

From: Mathy Vanhoef <[email protected]>

commit 965a7d72e798eb7af0aa67210e37cf7ecd1c9cad upstream.

Do not mix plaintext and encrypted fragments in protected Wi-Fi networks. This fixes CVE-2020-26147.

Previously, an attacker was able to first forward a legitimate encrypted fragment towards a victim, followed by a plaintext fragment. The encrypted and plaintext fragment would then be reassembled. For further details see Section 6.3 and Appendix D in the paper "Fragment and Forge:
Breaking Wi-Fi Through Frame Aggregation and Fragmentation".

Because of this change there are now two equivalent conditions in the code to determine if a received fragment requires sequential PNs, so we also move this test to a separate function to make the code easier to maintain.

Link https://lore.kernel.org/r/20210511200110.30c4394bb835.I5acfdb552cc1d20c339c262315950b3eac491397@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/mac80211/rx.c |   23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

net/mac80211/rx.c +12 ~~-11~~

@@ -2154,6 +2154,16 @@ ieee80211_reassemble_find(struct ieee802
 	return NULL;
 }
 
+static bool requires_sequential_pn(struct ieee80211_rx_data *rx, __le16 fc)
+{
+	return rx->key &&
+		(rx->key->conf.cipher == WLAN_CIPHER_SUITE_CCMP ||
+		 rx->key->conf.cipher == WLAN_CIPHER_SUITE_CCMP_256 ||
+		 rx->key->conf.cipher == WLAN_CIPHER_SUITE_GCMP ||
+		 rx->key->conf.cipher == WLAN_CIPHER_SUITE_GCMP_256) &&
+		ieee80211_has_protected(fc);
+}
+
 static ieee80211_rx_result debug_noinline
 ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx)
 {

@@ -2198,12 +2208,7 @@ ieee80211_rx_h_defragment(struct ieee802
 		/* This is the first fragment of a new frame. */
 		entry = ieee80211_reassemble_add(rx->sdata, frag, seq,
 						 rx->seqno_idx, &(rx->skb));
-		if (rx->key &&
-		    (rx->key->conf.cipher == WLAN_CIPHER_SUITE_CCMP ||
-		     rx->key->conf.cipher == WLAN_CIPHER_SUITE_CCMP_256 ||
-		     rx->key->conf.cipher == WLAN_CIPHER_SUITE_GCMP ||
-		     rx->key->conf.cipher == WLAN_CIPHER_SUITE_GCMP_256) &&
-		    ieee80211_has_protected(fc)) {
+		if (requires_sequential_pn(rx, fc)) {
 			int queue = rx->security_idx;
 
 			/* Store CCMP/GCMP PN so that we can verify that the

@@ -2245,11 +2250,7 @@ ieee80211_rx_h_defragment(struct ieee802
 		u8 pn[IEEE80211_CCMP_PN_LEN], *rpn;
 		int queue;
 
-		if (!rx->key ||
-		    (rx->key->conf.cipher != WLAN_CIPHER_SUITE_CCMP &&
-		     rx->key->conf.cipher != WLAN_CIPHER_SUITE_CCMP_256 &&
-		     rx->key->conf.cipher != WLAN_CIPHER_SUITE_GCMP &&
-		     rx->key->conf.cipher != WLAN_CIPHER_SUITE_GCMP_256))
+		if (!requires_sequential_pn(rx, fc))
 			return RX_DROP_UNUSABLE;
 		memcpy(pn, entry->last_pn, IEEE80211_CCMP_PN_LEN);
 		for (i = IEEE80211_CCMP_PN_LEN - 1; i >= 0; i--) {

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #15 lore raw

PATCH 15/177 mac80211: prevent mixed key and fragment cache attacks

From: Mathy Vanhoef <[email protected]>

commit 94034c40ab4a3fcf581fbc7f8fdf4e29943c4a24 upstream.

Simultaneously prevent mixed key attacks (CVE-2020-24587) and fragment cache attacks (CVE-2020-24586). This is accomplished by assigning a unique color to every key (per interface) and using this to track which key was used to decrypt a fragment. When reassembling frames, it is now checked whether all fragments were decrypted using the same key.

To assure that fragment cache attacks are also prevented, the ID that is assigned to keys is unique even over (re)associations and (re)connects. This means fragments separated by a (re)association or (re)connect will not be reassembled. Because mac80211 now also prevents the reassembly of mixed encrypted and plaintext fragments, all cache attacks are prevented.

Link https://lore.kernel.org/r/20210511200110.3f8290e59823.I622a67769ed39257327a362cfc09c812320eb979@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/mac80211/ieee80211_i.h |    1 +
 net/mac80211/key.c         |    7 +++++++
 net/mac80211/key.h         |    2 ++
 net/mac80211/rx.c          |    6 ++++++
 4 files changed, 16 insertions(+)

net/mac80211/ieee80211_i.h +1 -0

@@ -97,6 +97,7 @@ struct ieee80211_fragment_entry {
 	u8 rx_queue;
 	bool check_sequential_pn; /* needed for CCMP/GCMP */
 	u8 last_pn[6]; /* PN of the last fragment if CCMP was used */
+	unsigned int key_color;
 };

net/mac80211/key.c +7 -0

@@ -764,6 +764,7 @@ int ieee80211_key_link(struct ieee80211_
 		       struct ieee80211_sub_if_data *sdata,
 		       struct sta_info *sta)
 {
+	static atomic_t key_color = ATOMIC_INIT(0);
 	struct ieee80211_key *old_key;
 	int idx = key->conf.keyidx;
 	bool pairwise = key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE;

@@ -815,6 +816,12 @@ int ieee80211_key_link(struct ieee80211_
 	key->sdata = sdata;
 	key->sta = sta;
 
+	/*
+	 * Assign a unique ID to every key so we can easily prevent mixed
+	 * key and fragment cache attacks.
+	 */
+	key->color = atomic_inc_return(&key_color);
+
 	increment_tailroom_need_count(sdata);
 
 	ret = ieee80211_key_replace(sdata, sta, pairwise, old_key, key);

net/mac80211/key.h +2 -0

@@ -127,6 +127,8 @@ struct ieee80211_key {
 	} debugfs;
 #endif
 
+	unsigned int color;
+
 	/*
 	 * key config, must be last because it contains key
 	 * material as variable length member

net/mac80211/rx.c +6 -0

@@ -2215,6 +2215,7 @@ ieee80211_rx_h_defragment(struct ieee802
 			 * next fragment has a sequential PN value.
 			 */
 			entry->check_sequential_pn = true;
+			entry->key_color = rx->key->color;
 			memcpy(entry->last_pn,
 			       rx->key->u.ccmp.rx_pn[queue],
 			       IEEE80211_CCMP_PN_LEN);

@@ -2252,6 +2253,11 @@ ieee80211_rx_h_defragment(struct ieee802
 
 		if (!requires_sequential_pn(rx, fc))
 			return RX_DROP_UNUSABLE;
+
+		/* Prevent mixed key and fragment cache attacks */
+		if (entry->key_color != rx->key->color)
+			return RX_DROP_UNUSABLE;
+
 		memcpy(pn, entry->last_pn, IEEE80211_CCMP_PN_LEN);
 		for (i = IEEE80211_CCMP_PN_LEN - 1; i >= 0; i--) {
 			pn[i]++;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #16 lore raw

PATCH 16/177 mac80211: properly handle A-MSDUs that start with an RFC 1042 header

From: Mathy Vanhoef <[email protected]>

commit a1d5ff5651ea592c67054233b14b30bf4452999c upstream.

Properly parse A-MSDUs whose first 6 bytes happen to equal a rfc1042 header. This can occur in practice when the destination MAC address equals AA:AA:03:00:00:00. More importantly, this simplifies the next patch to mitigate A-MSDU injection attacks.

Link https://lore.kernel.org/r/20210511200110.0b2b886492f0.I23dd5d685fe16d3b0ec8106e8f01b59f499dffed@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 include/net/cfg80211.h |    4 ++--
 net/mac80211/rx.c      |    2 +-
 net/wireless/util.c    |    4 ++--
 3 files changed, 5 insertions(+), 5 deletions(-)

include/net/cfg80211.h +2 -2

@@ -5194,7 +5194,7 @@ unsigned int ieee80211_get_mesh_hdrlen(s
  */
 int ieee80211_data_to_8023_exthdr(struct sk_buff *skb, struct ethhdr *ehdr,
 				  const u8 *addr, enum nl80211_iftype iftype,
-				  u8 data_offset);
+				  u8 data_offset, bool is_amsdu);
 
 /**
  * ieee80211_data_to_8023 - convert an 802.11 data frame to 802.3

@@ -5206,7 +5206,7 @@ int ieee80211_data_to_8023_exthdr(struct
 static inline int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr,
 					 enum nl80211_iftype iftype)
 {
-	return ieee80211_data_to_8023_exthdr(skb, NULL, addr, iftype, 0);
+	return ieee80211_data_to_8023_exthdr(skb, NULL, addr, iftype, 0, false);
 }
 
 /**

net/mac80211/rx.c +1 -1

@@ -2634,7 +2634,7 @@ __ieee80211_rx_h_amsdu(struct ieee80211_
 	if (ieee80211_data_to_8023_exthdr(skb, &ethhdr,
 					  rx->sdata->vif.addr,
 					  rx->sdata->vif.type,
-					  data_offset))
+					  data_offset, true))
 		return RX_DROP_UNUSABLE;
 
 	ieee80211_amsdu_to_8023s(skb, &frame_list, dev->dev_addr,

net/wireless/util.c +2 -2

@@ -451,7 +451,7 @@ EXPORT_SYMBOL(ieee80211_get_mesh_hdrlen)
 
 int ieee80211_data_to_8023_exthdr(struct sk_buff *skb, struct ethhdr *ehdr,
 				  const u8 *addr, enum nl80211_iftype iftype,
-				  u8 data_offset)
+				  u8 data_offset, bool is_amsdu)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	struct {

@@ -539,7 +539,7 @@ int ieee80211_data_to_8023_exthdr(struct
 	skb_copy_bits(skb, hdrlen, &payload, sizeof(payload));
 	tmp.h_proto = payload.proto;
 
-	if (likely((ether_addr_equal(payload.hdr, rfc1042_header) &&
+	if (likely((!is_amsdu && ether_addr_equal(payload.hdr, rfc1042_header) &&
 		    tmp.h_proto != htons(ETH_P_AARP) &&
 		    tmp.h_proto != htons(ETH_P_IPX)) ||
 		   ether_addr_equal(payload.hdr, bridge_tunnel_header)))

Greg Kroah-Hartman

author maintainer

2021-05-31 14:20 UTC #17 lore raw

PATCH 17/177 cfg80211: mitigate A-MSDU aggregation attacks

From: Mathy Vanhoef <[email protected]>

commit 2b8a1fee3488c602aca8bea004a087e60806a5cf upstream.

Mitigate A-MSDU injection attacks (CVE-2020-24588) by detecting if the destination address of a subframe equals an RFC1042 (i.e., LLC/SNAP) header, and if so dropping the complete A-MSDU frame. This mitigates known attacks, although new (unknown) aggregation-based attacks may remain possible.

This defense works because in A-MSDU aggregation injection attacks, a normal encrypted Wi-Fi frame is turned into an A-MSDU frame. This means the first 6 bytes of the first A-MSDU subframe correspond to an RFC1042 header. In other words, the destination MAC address of the first A-MSDU subframe contains the start of an RFC1042 header during an aggregation attack. We can detect this and thereby prevent this specific attack. For details, see Section 7.2 of "Fragment and Forge: Breaking Wi-Fi Through Frame Aggregation and Fragmentation".

Note that for kernel 4.9 and above this patch depends on "mac80211:
properly handle A-MSDUs that start with a rfc1042 header". Otherwise this patch has no impact and attacks will remain possible.

Link https://lore.kernel.org/r/20210511200110.25d93176ddaf.I9e265b597f2cd23eb44573f35b625947b386a9de@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/wireless/util.c |    3 +++
 1 file changed, 3 insertions(+)

net/wireless/util.c +3 -0

@@ -681,6 +681,9 @@ void ieee80211_amsdu_to_8023s(struct sk_
 		remaining = skb->len - offset;
 		if (subframe_len > remaining)
 			goto purge;
+		/* mitigate A-MSDU aggregation injection attacks */
+		if (ether_addr_equal(eth.h_dest, rfc1042_header))
+			goto purge;
 
 		offset += sizeof(struct ethhdr);
 		last = remaining <= subframe_len + padding;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #18 lore raw

PATCH 18/177 mac80211: drop A-MSDUs on old ciphers

From: Johannes Berg <[email protected]>

commit 270032a2a9c4535799736142e1e7c413ca7b836e upstream.

With old ciphers (WEP and TKIP) we shouldn't be using A-MSDUs since A-MSDUs are only supported if we know that they are, and the only practical way for that is HT support which doesn't support old ciphers.

However, we would normally accept them anyway. Since we check the MMIC before deaggregating A-MSDUs, and the A-MSDU bit in the QoS header is not protected in TKIP (or WEP), this enables attacks similar to CVE-2020-24588. To prevent that, drop A-MSDUs completely with old ciphers.

Link https://lore.kernel.org/r/20210511200110.076543300172.I548e6e71f1ee9cad4b9a37bf212ae7db723587aa@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/mac80211/rx.c |   19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

net/mac80211/rx.c +18 -1

@@ -6,7 +6,7 @@
  * Copyright 2007-2010	Johannes Berg <[email protected]>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
  * Copyright(c) 2015 - 2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2019 Intel Corporation
+ * Copyright (C) 2018-2021 Intel Corporation
  */
 
 #include <linux/jiffies.h>

@@ -2691,6 +2691,23 @@ ieee80211_rx_h_amsdu(struct ieee80211_rx
 	if (is_multicast_ether_addr(hdr->addr1))
 		return RX_DROP_UNUSABLE;
 
+	if (rx->key) {
+		/*
+		 * We should not receive A-MSDUs on pre-HT connections,
+		 * and HT connections cannot use old ciphers. Thus drop
+		 * them, as in those cases we couldn't even have SPP
+		 * A-MSDUs or such.
+		 */
+		switch (rx->key->conf.cipher) {
+		case WLAN_CIPHER_SUITE_WEP40:
+		case WLAN_CIPHER_SUITE_WEP104:
+		case WLAN_CIPHER_SUITE_TKIP:
+			return RX_DROP_UNUSABLE;
+		default:
+			break;
+		}
+	}
+
 	return __ieee80211_rx_h_amsdu(rx, 0);
 }

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #19 lore raw

PATCH 19/177 mac80211: add fragment cache to sta_info

From: Johannes Berg <[email protected]>

commit 3a11ce08c45b50d69c891d71760b7c5b92074709 upstream.

Prior patches protected against fragmentation cache attacks by coloring keys, but this shows that it can lead to issues when multiple stations use the same sequence number. Add a fragment cache to struct sta_info (in addition to the one in the interface) to separate fragments for different stations properly.

This then automatically clear most of the fragment cache when a station disconnects (or reassociates) from an AP, or when client interfaces disconnect from the network, etc.

On the way, also fix the comment there since this brings us in line with the recommendation in 802.11-2016 ("An AP should support ..."). Additionally, remove a useless condition (since there's no problem purging an already empty list).

Link https://lore.kernel.org/r/20210511200110.fc35046b0d52.I1ef101e3784d13e8f6600d83de7ec9a3a45bcd52@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/mac80211/ieee80211_i.h |   26 ++++----------------------
 net/mac80211/iface.c       |   11 +++--------
 net/mac80211/rx.c          |   41 ++++++++++++++++++++++++++++++-----------
 net/mac80211/sta_info.c    |    6 +++++-
 net/mac80211/sta_info.h    |   31 +++++++++++++++++++++++++++++++
 5 files changed, 73 insertions(+), 42 deletions(-)

net/mac80211/ieee80211_i.h +4 ~~-22~~

@@ -50,12 +50,6 @@ struct ieee80211_local;
 #define IEEE80211_ENCRYPT_HEADROOM 8
 #define IEEE80211_ENCRYPT_TAILROOM 18
 
-/* IEEE 802.11 (Ch. 9.5 Defragmentation) requires support for concurrent
- * reception of at least three fragmented frames. This limit can be increased
- * by changing this define, at the cost of slower frame reassembly and
- * increased memory use (about 2 kB of RAM per entry). */
-#define IEEE80211_FRAGMENT_MAX 4
-
 /* power level hasn't been configured (or set to automatic) */
 #define IEEE80211_UNSET_POWER_LEVEL	INT_MIN

@@ -88,19 +82,6 @@ extern const u8 ieee80211_ac_to_qos_mask
 
 #define IEEE80211_MAX_NAN_INSTANCE_ID 255
 
-struct ieee80211_fragment_entry {
-	struct sk_buff_head skb_list;
-	unsigned long first_frag_time;
-	u16 seq;
-	u16 extra_len;
-	u16 last_frag;
-	u8 rx_queue;
-	bool check_sequential_pn; /* needed for CCMP/GCMP */
-	u8 last_pn[6]; /* PN of the last fragment if CCMP was used */
-	unsigned int key_color;
-};
-
-
 struct ieee80211_bss {
 	u32 device_ts_beacon, device_ts_presp;

@@ -895,9 +876,7 @@ struct ieee80211_sub_if_data {
 
 	char name[IFNAMSIZ];
 
-	/* Fragment table for host-based reassembly */
-	struct ieee80211_fragment_entry	fragments[IEEE80211_FRAGMENT_MAX];
-	unsigned int fragment_next;
+	struct ieee80211_fragment_cache frags;
 
 	/* TID bitmap for NoAck policy */
 	u16 noack_map;

@@ -2257,4 +2236,7 @@ extern const struct ethtool_ops ieee8021
 #define debug_noinline
 #endif
 
+void ieee80211_init_frag_cache(struct ieee80211_fragment_cache *cache);
+void ieee80211_destroy_frag_cache(struct ieee80211_fragment_cache *cache);
+
 #endif /* IEEE80211_I_H */

net/mac80211/iface.c +3 -8

@@ -8,7 +8,7 @@
  * Copyright 2008, Johannes Berg <[email protected]>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
  * Copyright (c) 2016        Intel Deutschland GmbH
- * Copyright (C) 2018 Intel Corporation
+ * Copyright (C) 2018-2021 Intel Corporation
  */
 #include <linux/slab.h>
 #include <linux/kernel.h>

@@ -1108,16 +1108,12 @@ static void ieee80211_set_multicast_list
  */
 static void ieee80211_teardown_sdata(struct ieee80211_sub_if_data *sdata)
 {
-	int i;
-
 	/* free extra data */
 	ieee80211_free_keys(sdata, false);
 
 	ieee80211_debugfs_remove_netdev(sdata);
 
-	for (i = 0; i < IEEE80211_FRAGMENT_MAX; i++)
-		__skb_queue_purge(&sdata->fragments[i].skb_list);
-	sdata->fragment_next = 0;
+	ieee80211_destroy_frag_cache(&sdata->frags);
 
 	if (ieee80211_vif_is_mesh(&sdata->vif))
 		ieee80211_mesh_teardown_sdata(sdata);

@@ -1827,8 +1823,7 @@ int ieee80211_if_add(struct ieee80211_lo
 	sdata->wdev.wiphy = local->hw.wiphy;
 	sdata->local = local;
 
-	for (i = 0; i < IEEE80211_FRAGMENT_MAX; i++)
-		skb_queue_head_init(&sdata->fragments[i].skb_list);
+	ieee80211_init_frag_cache(&sdata->frags);
 
 	INIT_LIST_HEAD(&sdata->key_list);

net/mac80211/rx.c +30 ~~-11~~

96 lines not on this page · raw message

net/mac80211/sta_info.c +5 -1

@@ -4,7 +4,7 @@
  * Copyright 2006-2007	Jiri Benc <[email protected]>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
  * Copyright (C) 2015 - 2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2020 Intel Corporation
+ * Copyright (C) 2018-2021 Intel Corporation
  */
 
 #include <linux/module.h>

@@ -378,6 +378,8 @@ struct sta_info *sta_info_alloc(struct i
 
 	u64_stats_init(&sta->rx_stats.syncp);
 
+	ieee80211_init_frag_cache(&sta->frags);
+
 	sta->sta_state = IEEE80211_STA_NONE;
 
 	/* Mark TID as unreserved */

@@ -1085,6 +1087,8 @@ static void __sta_info_destroy_part2(str
 
 	ieee80211_sta_debugfs_remove(sta);
 
+	ieee80211_destroy_frag_cache(&sta->frags);
+
 	cleanup_single_sta(sta);
 }

net/mac80211/sta_info.h +31 -0

@@ -3,6 +3,7 @@
  * Copyright 2002-2005, Devicescape Software, Inc.
  * Copyright 2013-2014  Intel Mobile Communications GmbH
  * Copyright(c) 2015-2017 Intel Deutschland GmbH
+ * Copyright(c) 2020-2021 Intel Corporation
  */
 
 #ifndef STA_INFO_H

@@ -426,6 +427,33 @@ struct ieee80211_sta_rx_stats {
 };
 
 /*
+ * IEEE 802.11-2016 (10.6 "Defragmentation") recommends support for "concurrent
+ * reception of at least one MSDU per access category per associated STA"
+ * on APs, or "at least one MSDU per access category" on other interface types.
+ *
+ * This limit can be increased by changing this define, at the cost of slower
+ * frame reassembly and increased memory use while fragments are pending.
+ */
+#define IEEE80211_FRAGMENT_MAX 4
+
+struct ieee80211_fragment_entry {
+	struct sk_buff_head skb_list;
+	unsigned long first_frag_time;
+	u16 seq;
+	u16 extra_len;
+	u16 last_frag;
+	u8 rx_queue;
+	bool check_sequential_pn; /* needed for CCMP/GCMP */
+	u8 last_pn[6]; /* PN of the last fragment if CCMP was used */
+	unsigned int key_color;
+};
+
+struct ieee80211_fragment_cache {
+	struct ieee80211_fragment_entry	entries[IEEE80211_FRAGMENT_MAX];
+	unsigned int next;
+};
+
+/*
  * The bandwidth threshold below which the per-station CoDel parameters will be
  * scaled to be more lenient (to prevent starvation of slow stations). This
  * value will be scaled by the number of active stations when it is being

@@ -518,6 +546,7 @@ struct ieee80211_sta_rx_stats {
  * @status_stats.last_ack_signal: last ACK signal
  * @status_stats.ack_signal_filled: last ACK signal validity
  * @status_stats.avg_ack_signal: average ACK signal
+ * @frags: fragment cache
  */
 struct sta_info {
 	/* General information, mostly static */

@@ -623,6 +652,8 @@ struct sta_info {
 
 	struct cfg80211_chan_def tdls_chandef;
 
+	struct ieee80211_fragment_cache frags;
+
 	/* keep last! */
 	struct ieee80211_sta sta;
 };

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #20 lore raw

PATCH 20/177 mac80211: check defrag PN against current frame

From: Johannes Berg <[email protected]>

commit bf30ca922a0c0176007e074b0acc77ed345e9990 upstream.

As pointed out by Mathy Vanhoef, we implement the RX PN check on fragmented frames incorrectly - we check against the last received PN prior to the new frame, rather than to the one in this frame itself.

Prior patches addressed the security issue here, but in order to be able to reason better about the code, fix it to really compare against the current frame's PN, not the last stored one.

Link https://lore.kernel.org/r/20210511200110.bfbc340ff071.Id0b690e581da7d03d76df90bb0e3fd55930bc8a0@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/mac80211/ieee80211_i.h |   11 +++++++++--
 net/mac80211/rx.c          |    5 ++---
 net/mac80211/wpa.c         |   13 +++++++++----
 3 files changed, 20 insertions(+), 9 deletions(-)

net/mac80211/ieee80211_i.h +9 -2

@@ -222,8 +222,15 @@ struct ieee80211_rx_data {
 	 */
 	int security_idx;
 
-	u32 tkip_iv32;
-	u16 tkip_iv16;
+	union {
+		struct {
+			u32 iv32;
+			u16 iv16;
+		} tkip;
+		struct {
+			u8 pn[IEEE80211_CCMP_PN_LEN];
+		} ccm_gcm;
+	};
 };
 
 struct ieee80211_csa_settings {

net/mac80211/rx.c +2 -3

@@ -2268,7 +2268,6 @@ ieee80211_rx_h_defragment(struct ieee802
 	if (entry->check_sequential_pn) {
 		int i;
 		u8 pn[IEEE80211_CCMP_PN_LEN], *rpn;
-		int queue;
 
 		if (!requires_sequential_pn(rx, fc))
 			return RX_DROP_UNUSABLE;

@@ -2283,8 +2282,8 @@ ieee80211_rx_h_defragment(struct ieee802
 			if (pn[i])
 				break;
 		}
-		queue = rx->security_idx;
-		rpn = rx->key->u.ccmp.rx_pn[queue];
+
+		rpn = rx->ccm_gcm.pn;
 		if (memcmp(pn, rpn, IEEE80211_CCMP_PN_LEN))
 			return RX_DROP_UNUSABLE;
 		memcpy(entry->last_pn, pn, IEEE80211_CCMP_PN_LEN);

net/mac80211/wpa.c +9 -4

@@ -3,6 +3,7 @@
  * Copyright 2002-2004, Instant802 Networks, Inc.
  * Copyright 2008, Jouni Malinen <[email protected]>
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
+ * Copyright (C) 2020-2021 Intel Corporation
  */
 
 #include <linux/netdevice.h>

@@ -167,8 +168,8 @@ ieee80211_rx_h_michael_mic_verify(struct
 
 update_iv:
 	/* update IV in key information to be able to detect replays */
-	rx->key->u.tkip.rx[rx->security_idx].iv32 = rx->tkip_iv32;
-	rx->key->u.tkip.rx[rx->security_idx].iv16 = rx->tkip_iv16;
+	rx->key->u.tkip.rx[rx->security_idx].iv32 = rx->tkip.iv32;
+	rx->key->u.tkip.rx[rx->security_idx].iv16 = rx->tkip.iv16;
 
 	return RX_CONTINUE;

@@ -294,8 +295,8 @@ ieee80211_crypto_tkip_decrypt(struct iee
 					  key, skb->data + hdrlen,
 					  skb->len - hdrlen, rx->sta->sta.addr,
 					  hdr->addr1, hwaccel, rx->security_idx,
-					  &rx->tkip_iv32,
-					  &rx->tkip_iv16);
+					  &rx->tkip.iv32,
+					  &rx->tkip.iv16);
 	if (res != TKIP_DECRYPT_OK)
 		return RX_DROP_UNUSABLE;

@@ -553,6 +554,8 @@ ieee80211_crypto_ccmp_decrypt(struct iee
 		}
 
 		memcpy(key->u.ccmp.rx_pn[queue], pn, IEEE80211_CCMP_PN_LEN);
+		if (unlikely(ieee80211_is_frag(hdr)))
+			memcpy(rx->ccm_gcm.pn, pn, IEEE80211_CCMP_PN_LEN);
 	}
 
 	/* Remove CCMP header and MIC */

@@ -781,6 +784,8 @@ ieee80211_crypto_gcmp_decrypt(struct iee
 		}
 
 		memcpy(key->u.gcmp.rx_pn[queue], pn, IEEE80211_GCMP_PN_LEN);
+		if (unlikely(ieee80211_is_frag(hdr)))
+			memcpy(rx->ccm_gcm.pn, pn, IEEE80211_CCMP_PN_LEN);
 	}
 
 	/* Remove GCMP header and MIC */

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #21 lore raw

PATCH 21/177 mac80211: prevent attacks on TKIP/WEP as well

From: Johannes Berg <[email protected]>

commit 7e44a0b597f04e67eee8cdcbe7ee706c6f5de38b upstream.

Similar to the issues fixed in previous patches, TKIP and WEP should be protected even if for TKIP we have the Michael MIC protecting it, and WEP is broken anyway.

However, this also somewhat protects potential other algorithms that drivers might implement.

Link https://lore.kernel.org/r/20210511200110.430e8c202313.Ia37e4e5b6b3eaab1a5ae050e015f6c92859dbe27@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/mac80211/rx.c       |   12 ++++++++++++
 net/mac80211/sta_info.h |    3 ++-
 2 files changed, 14 insertions(+), 1 deletion(-)

net/mac80211/rx.c +12 -0

@@ -2234,6 +2234,7 @@ ieee80211_rx_h_defragment(struct ieee802
 			 * next fragment has a sequential PN value.
 			 */
 			entry->check_sequential_pn = true;
+			entry->is_protected = true;
 			entry->key_color = rx->key->color;
 			memcpy(entry->last_pn,
 			       rx->key->u.ccmp.rx_pn[queue],

@@ -2246,6 +2247,9 @@ ieee80211_rx_h_defragment(struct ieee802
 				     sizeof(rx->key->u.gcmp.rx_pn[queue]));
 			BUILD_BUG_ON(IEEE80211_CCMP_PN_LEN !=
 				     IEEE80211_GCMP_PN_LEN);
+		} else if (rx->key && ieee80211_has_protected(fc)) {
+			entry->is_protected = true;
+			entry->key_color = rx->key->color;
 		}
 		return RX_QUEUED;
 	}

@@ -2287,6 +2291,14 @@ ieee80211_rx_h_defragment(struct ieee802
 		if (memcmp(pn, rpn, IEEE80211_CCMP_PN_LEN))
 			return RX_DROP_UNUSABLE;
 		memcpy(entry->last_pn, pn, IEEE80211_CCMP_PN_LEN);
+	} else if (entry->is_protected &&
+		   (!rx->key || !ieee80211_has_protected(fc) ||
+		    rx->key->color != entry->key_color)) {
+		/* Drop this as a mixed key or fragment cache attack, even
+		 * if for TKIP Michael MIC should protect us, and WEP is a
+		 * lost cause anyway.
+		 */
+		return RX_DROP_UNUSABLE;
 	}
 
 	skb_pull(rx->skb, ieee80211_hdrlen(fc));

net/mac80211/sta_info.h +2 -1

@@ -443,7 +443,8 @@ struct ieee80211_fragment_entry {
 	u16 extra_len;
 	u16 last_frag;
 	u8 rx_queue;
-	bool check_sequential_pn; /* needed for CCMP/GCMP */
+	u8 check_sequential_pn:1, /* needed for CCMP/GCMP */
+	   is_protected:1;
 	u8 last_pn[6]; /* PN of the last fragment if CCMP was used */
 	unsigned int key_color;
 };

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #22 lore raw

PATCH 22/177 mac80211: do not accept/forward invalid EAPOL frames

From: Johannes Berg <[email protected]>

commit a8c4d76a8dd4fb9666fc8919a703d85fb8f44ed8 upstream.

EAPOL frames are used for authentication and key management between the AP and each individual STA associated in the BSS. Those frames are not supposed to be sent by one associated STA to another associated STA (either unicast for broadcast/multicast).

Similarly, in 802.11 they're supposed to be sent to the authenticator (AP) address.

Since it is possible for unexpected EAPOL frames to result in misbehavior in supplicant implementations, it is better for the AP to not allow such cases to be forwarded to other clients either directly, or indirectly if the AP interface is part of a bridge.

Accept EAPOL (control port) frames only if they're transmitted to the own address, or, due to interoperability concerns, to the PAE group address.

Disable forwarding of EAPOL (or well, the configured control port protocol) frames back to wireless medium in all cases. Previously, these frames were accepted from fully authenticated and authorized stations and also from unauthenticated stations for one of the cases.

Additionally, to avoid forwarding by the bridge, rewrite the PAE group address case to the local MAC address.

Co-developed-by Jouni Malinen <[email protected]>

Link https://lore.kernel.org/r/20210511200110.cb327ed0cabe.Ib7dcffa2a31f0913d660de65ba3c8aca75b1d10f@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/mac80211/rx.c |   33 +++++++++++++++++++++++++++------
 1 file changed, 27 insertions(+), 6 deletions(-)

net/mac80211/rx.c +27 -6

@@ -2484,13 +2484,13 @@ static bool ieee80211_frame_allowed(stru
 	struct ethhdr *ehdr = (struct ethhdr *) rx->skb->data;
 
 	/*
-	 * Allow EAPOL frames to us/the PAE group address regardless
-	 * of whether the frame was encrypted or not.
+	 * Allow EAPOL frames to us/the PAE group address regardless of
+	 * whether the frame was encrypted or not, and always disallow
+	 * all other destination addresses for them.
 	 */
-	if (ehdr->h_proto == rx->sdata->control_port_protocol &&
-	    (ether_addr_equal(ehdr->h_dest, rx->sdata->vif.addr) ||
-	     ether_addr_equal(ehdr->h_dest, pae_group_addr)))
-		return true;
+	if (unlikely(ehdr->h_proto == rx->sdata->control_port_protocol))
+		return ether_addr_equal(ehdr->h_dest, rx->sdata->vif.addr) ||
+		       ether_addr_equal(ehdr->h_dest, pae_group_addr);
 
 	if (ieee80211_802_1x_port_control(rx) ||
 	    ieee80211_drop_unencrypted(rx, fc))

@@ -2514,8 +2514,28 @@ static void ieee80211_deliver_skb_to_loc
 		cfg80211_rx_control_port(dev, skb, noencrypt);
 		dev_kfree_skb(skb);
 	} else {
+		struct ethhdr *ehdr = (void *)skb_mac_header(skb);
+
 		memset(skb->cb, 0, sizeof(skb->cb));
 
+		/*
+		 * 802.1X over 802.11 requires that the authenticator address
+		 * be used for EAPOL frames. However, 802.1X allows the use of
+		 * the PAE group address instead. If the interface is part of
+		 * a bridge and we pass the frame with the PAE group address,
+		 * then the bridge will forward it to the network (even if the
+		 * client was not associated yet), which isn't supposed to
+		 * happen.
+		 * To avoid that, rewrite the destination address to our own
+		 * address, so that the authenticator (e.g. hostapd) will see
+		 * the frame, but bridge won't forward it anywhere else. Note
+		 * that due to earlier filtering, the only other address can
+		 * be the PAE group address.
+		 */
+		if (unlikely(skb->protocol == sdata->control_port_protocol &&
+			     !ether_addr_equal(ehdr->h_dest, sdata->vif.addr)))
+			ether_addr_copy(ehdr->h_dest, sdata->vif.addr);
+
 		/* deliver to local stack */
 		if (rx->napi)
 			napi_gro_receive(rx->napi, skb);

@@ -2555,6 +2575,7 @@ ieee80211_deliver_skb(struct ieee80211_r
 	if ((sdata->vif.type == NL80211_IFTYPE_AP ||
 	     sdata->vif.type == NL80211_IFTYPE_AP_VLAN) &&
 	    !(sdata->flags & IEEE80211_SDATA_DONT_BRIDGE_PACKETS) &&
+	    ehdr->h_proto != rx->sdata->control_port_protocol &&
 	    (sdata->vif.type != NL80211_IFTYPE_AP_VLAN || !sdata->u.vlan.sta)) {
 		if (is_multicast_ether_addr(ehdr->h_dest) &&
 		    ieee80211_vif_get_num_mcast_if(sdata) != 0) {

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #23 lore raw

PATCH 23/177 mac80211: extend protection against mixed key and fragment cache attacks

From: Wen Gong <[email protected]>

commit 3edc6b0d6c061a70d8ca3c3c72eb1f58ce29bfb1 upstream.

For some chips/drivers, e.g., QCA6174 with ath10k, the decryption is done by the hardware, and the Protected bit in the Frame Control field is cleared in the lower level driver before the frame is passed to mac80211. In such cases, the condition for ieee80211_has_protected() is not met in ieee80211_rx_h_defragment() of mac80211 and the new security validation steps are not executed.

Extend mac80211 to cover the case where the Protected bit has been cleared, but the frame is indicated as having been decrypted by the hardware. This extends protection against mixed key and fragment cache attack for additional drivers/chips. This fixes CVE-2020-24586 and CVE-2020-24587 for such cases.

Tested-on: QCA6174 hw3.2 PCI WLAN.RM.4.4.1-00110-QCARMSWP-1

Signed-off-by Wen Gong <[email protected]>

Link https://lore.kernel.org/r/20210511200110.037aa5ca0390.I7bb888e2965a0db02a67075fcb5deb50eb7408aa@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/mac80211/rx.c |   13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

net/mac80211/rx.c +11 -2

@@ -2189,6 +2189,7 @@ ieee80211_rx_h_defragment(struct ieee802
 	unsigned int frag, seq;
 	struct ieee80211_fragment_entry *entry;
 	struct sk_buff *skb;
+	struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb);
 
 	hdr = (struct ieee80211_hdr *)rx->skb->data;
 	fc = hdr->frame_control;

@@ -2247,7 +2248,9 @@ ieee80211_rx_h_defragment(struct ieee802
 				     sizeof(rx->key->u.gcmp.rx_pn[queue]));
 			BUILD_BUG_ON(IEEE80211_CCMP_PN_LEN !=
 				     IEEE80211_GCMP_PN_LEN);
-		} else if (rx->key && ieee80211_has_protected(fc)) {
+		} else if (rx->key &&
+			   (ieee80211_has_protected(fc) ||
+			    (status->flag & RX_FLAG_DECRYPTED))) {
 			entry->is_protected = true;
 			entry->key_color = rx->key->color;
 		}

@@ -2292,13 +2295,19 @@ ieee80211_rx_h_defragment(struct ieee802
 			return RX_DROP_UNUSABLE;
 		memcpy(entry->last_pn, pn, IEEE80211_CCMP_PN_LEN);
 	} else if (entry->is_protected &&
-		   (!rx->key || !ieee80211_has_protected(fc) ||
+		   (!rx->key ||
+		    (!ieee80211_has_protected(fc) &&
+		     !(status->flag & RX_FLAG_DECRYPTED)) ||
 		    rx->key->color != entry->key_color)) {
 		/* Drop this as a mixed key or fragment cache attack, even
 		 * if for TKIP Michael MIC should protect us, and WEP is a
 		 * lost cause anyway.
 		 */
 		return RX_DROP_UNUSABLE;
+	} else if (entry->is_protected && rx->key &&
+		   entry->key_color != rx->key->color &&
+		   (status->flag & RX_FLAG_DECRYPTED)) {
+		return RX_DROP_UNUSABLE;
 	}
 
 	skb_pull(rx->skb, ieee80211_hdrlen(fc));

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #24 lore raw

PATCH 24/177 ath10k: add CCMP PN replay protection for fragmented frames for PCIe

From: Wen Gong <[email protected]>

commit a1166b2653db2f3de7338b9fb8a0f6e924b904ee upstream.

PN replay check for not fragmented frames is finished in the firmware, but this was not done for fragmented frames when ath10k is used with QCA6174/QCA6377 PCIe. mac80211 has the function ieee80211_rx_h_defragment() for PN replay check for fragmented frames, but this does not get checked with QCA6174 due to the ieee80211_has_protected() condition not matching the cleared Protected bit case.

Validate the PN of received fragmented frames within ath10k when CCMP is used and drop the fragment if the PN is not correct (incremented by exactly one from the previous fragment). This applies only for QCA6174/QCA6377 PCIe.

Tested-on: QCA6174 hw3.2 PCI WLAN.RM.4.4.1-00110-QCARMSWP-1

Signed-off-by Wen Gong <[email protected]>

Link https://lore.kernel.org/r/20210511200110.9ba2664866a4.I756e47b67e210dba69966d989c4711ffc02dc6bc@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/wireless/ath/ath10k/htt.h    |    1 
 drivers/net/wireless/ath/ath10k/htt_rx.c |   99 +++++++++++++++++++++++++++++--
 2 files changed, 96 insertions(+), 4 deletions(-)

drivers/net/wireless/ath/ath10k/htt.h +1 -0

@@ -835,6 +835,7 @@ enum htt_security_types {
 
 #define ATH10K_HTT_TXRX_PEER_SECURITY_MAX 2
 #define ATH10K_TXRX_NUM_EXT_TIDS 19
+#define ATH10K_TXRX_NON_QOS_TID 16
 
 enum htt_security_flags {
 #define HTT_SECURITY_TYPE_MASK 0x7F

drivers/net/wireless/ath/ath10k/htt_rx.c +95 -4

137 lines not on this page · raw message

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #25 lore raw

PATCH 25/177 ath10k: drop fragments with multicast DA for PCIe

From: Wen Gong <[email protected]>

commit 65c415a144ad8132b6a6d97d4a1919ffc728e2d1 upstream.

Fragmentation is not used with multicast frames. Discard unexpected fragments with multicast DA. This fixes CVE-2020-26145.

Tested-on: QCA6174 hw3.2 PCI WLAN.RM.4.4.1-00110-QCARMSWP-1

Signed-off-by Wen Gong <[email protected]>

Link https://lore.kernel.org/r/20210511200110.5a0bd289bda8.Idd6ebea20038fb1cfee6de924aa595e5647c9eae@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/wireless/ath/ath10k/htt_rx.c |   23 ++++++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)

drivers/net/wireless/ath/ath10k/htt_rx.c +20 -3

@@ -1761,6 +1761,16 @@ static u64 ath10k_htt_rx_h_get_pn(struct
 	return pn;
 }
 
+static bool ath10k_htt_rx_h_frag_multicast_check(struct ath10k *ar,
+						 struct sk_buff *skb,
+						 u16 offset)
+{
+	struct ieee80211_hdr *hdr;
+
+	hdr = (struct ieee80211_hdr *)(skb->data + offset);
+	return !is_multicast_ether_addr(hdr->addr1);
+}
+
 static bool ath10k_htt_rx_h_frag_pn_check(struct ath10k *ar,
 					  struct sk_buff *skb,
 					  u16 peer_id,

@@ -1832,7 +1842,7 @@ static void ath10k_htt_rx_h_mpdu(struct
 	bool is_decrypted;
 	bool is_mgmt;
 	u32 attention;
-	bool frag_pn_check = true;
+	bool frag_pn_check = true, multicast_check = true;
 
 	if (skb_queue_empty(amsdu))
 		return;

@@ -1939,13 +1949,20 @@ static void ath10k_htt_rx_h_mpdu(struct
 								      0,
 								      enctype);
 
-		if (!frag_pn_check) {
-			/* Discard the fragment with invalid PN */
+		if (frag)
+			multicast_check = ath10k_htt_rx_h_frag_multicast_check(ar,
+									       msdu,
+									       0);
+
+		if (!frag_pn_check || !multicast_check) {
+			/* Discard the fragment with invalid PN or multicast DA
+			 */
 			temp = msdu->prev;
 			__skb_unlink(msdu, amsdu);
 			dev_kfree_skb_any(msdu);
 			msdu = temp;
 			frag_pn_check = true;
+			multicast_check = true;
 			continue;
 		}

Greg Kroah-Hartman

author maintainer

2021-05-31 14:14 UTC #26 lore raw

PATCH 26/177 ath10k: drop fragments with multicast DA for SDIO

From: Wen Gong <[email protected]>

commit 40e7462dad6f3d06efdb17d26539e61ab6e34db1 upstream.

Fragmentation is not used with multicast frames. Discard unexpected fragments with multicast DA. This fixes CVE-2020-26145.

Tested-on: QCA6174 hw3.2 SDIO WLAN.RMH.4.4.1-00049

Signed-off-by Wen Gong <[email protected]>

Link https://lore.kernel.org/r/20210511200110.9ca6ca7945a9.I1e18b514590af17c155bda86699bc3a971a8dcf4@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/wireless/ath/ath10k/htt_rx.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

drivers/net/wireless/ath/ath10k/htt_rx.c +7 -2

@@ -2605,6 +2605,13 @@ static bool ath10k_htt_rx_proc_rx_frag_i
 	rx_desc = (struct htt_hl_rx_desc *)(skb->data + tot_hdr_len);
 	rx_desc_info = __le32_to_cpu(rx_desc->info);
 
+	hdr = (struct ieee80211_hdr *)((u8 *)rx_desc + rx_hl->fw_desc.len);
+
+	if (is_multicast_ether_addr(hdr->addr1)) {
+		/* Discard the fragment with multicast DA */
+		goto err;
+	}
+
 	if (!MS(rx_desc_info, HTT_RX_DESC_HL_INFO_ENCRYPTED)) {
 		spin_unlock_bh(&ar->data_lock);
 		return ath10k_htt_rx_proc_rx_ind_hl(htt, &resp->rx_ind_hl, skb,

@@ -2612,8 +2619,6 @@ static bool ath10k_htt_rx_proc_rx_frag_i
 						    HTT_RX_NON_TKIP_MIC);
 	}
 
-	hdr = (struct ieee80211_hdr *)((u8 *)rx_desc + rx_hl->fw_desc.len);
-
 	if (ieee80211_has_retry(hdr->frame_control))
 		goto err;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #27 lore raw

PATCH 27/177 ath10k: drop MPDU which has discard flag set by firmware for SDIO

From: Wen Gong <[email protected]>

commit 079a108feba474b4b32bd3471db03e11f2f83b81 upstream.

When the discard flag is set by the firmware for an MPDU, it should be dropped. This allows a mitigation for CVE-2020-24588 to be implemented in the firmware.

Tested-on: QCA6174 hw3.2 SDIO WLAN.RMH.4.4.1-00049

Signed-off-by Wen Gong <[email protected]>

Link https://lore.kernel.org/r/20210511200110.11968c725b5c.Idd166365ebea2771c0c0a38c78b5060750f90e17@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/wireless/ath/ath10k/htt_rx.c  |    5 +++++
 drivers/net/wireless/ath/ath10k/rx_desc.h |   14 +++++++++++++-
 2 files changed, 18 insertions(+), 1 deletion(-)

drivers/net/wireless/ath/ath10k/htt_rx.c +5 -0

@@ -2305,6 +2305,11 @@ static bool ath10k_htt_rx_proc_rx_ind_hl
 	fw_desc = &rx->fw_desc;
 	rx_desc_len = fw_desc->len;
 
+	if (fw_desc->u.bits.discard) {
+		ath10k_dbg(ar, ATH10K_DBG_HTT, "htt discard mpdu\n");
+		goto err;
+	}
+
 	/* I have not yet seen any case where num_mpdu_ranges > 1.
 	 * qcacld does not seem handle that case either, so we introduce the
 	 * same limitiation here as well.

drivers/net/wireless/ath/ath10k/rx_desc.h +13 -1

@@ -1282,7 +1282,19 @@ struct fw_rx_desc_base {
 #define FW_RX_DESC_UDP              (1 << 6)
 
 struct fw_rx_desc_hl {
-	u8 info0;
+	union {
+		struct {
+		u8 discard:1,
+		   forward:1,
+		   any_err:1,
+		   dup_err:1,
+		   reserved:1,
+		   inspect:1,
+		   extension:2;
+		} bits;
+		u8 info0;
+	} u;
+
 	u8 version;
 	u8 len;
 	u8 flags;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #28 lore raw

PATCH 28/177 ath10k: Fix TKIP Michael MIC verification for PCIe

From: Wen Gong <[email protected]>

commit 0dc267b13f3a7e8424a898815dd357211b737330 upstream.

TKIP Michael MIC was not verified properly for PCIe cases since the validation steps in ieee80211_rx_h_michael_mic_verify() in mac80211 did not get fully executed due to unexpected flag values in ieee80211_rx_status.

Fix this by setting the flags property to meet mac80211 expectations for performing Michael MIC validation there. This fixes CVE-2020-26141. It does the same as ath10k_htt_rx_proc_rx_ind_hl() for SDIO which passed MIC verification case. This applies only to QCA6174/QCA9377 PCIe.

Tested-on: QCA6174 hw3.2 PCI WLAN.RM.4.4.1-00110-QCARMSWP-1

Signed-off-by Wen Gong <[email protected]>

Link https://lore.kernel.org/r/20210511200110.c3f1d42c6746.I795593fcaae941c471425b8c7d5f7bb185d29142@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/wireless/ath/ath10k/htt_rx.c |   10 ++++++++++
 1 file changed, 10 insertions(+)

drivers/net/wireless/ath/ath10k/htt_rx.c +10 -0

@@ -1967,6 +1967,11 @@ static void ath10k_htt_rx_h_mpdu(struct
 		}
 
 		ath10k_htt_rx_h_csum_offload(msdu);
+
+		if (frag && !fill_crypt_header &&
+		    enctype == HTT_RX_MPDU_ENCRYPT_TKIP_WPA)
+			status->flag &= ~RX_FLAG_MMIC_STRIPPED;
+
 		ath10k_htt_rx_h_undecap(ar, msdu, status, first_hdr, enctype,
 					is_decrypted);

@@ -1984,6 +1989,11 @@ static void ath10k_htt_rx_h_mpdu(struct
 
 		hdr = (void *)msdu->data;
 		hdr->frame_control &= ~__cpu_to_le16(IEEE80211_FCTL_PROTECTED);
+
+		if (frag && !fill_crypt_header &&
+		    enctype == HTT_RX_MPDU_ENCRYPT_TKIP_WPA)
+			status->flag &= ~RX_FLAG_IV_STRIPPED &
+					~RX_FLAG_MMIC_STRIPPED;
 	}
 }

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #29 lore raw

PATCH 29/177 ath10k: Validate first subframe of A-MSDU before processing the list

From: Sriram R <[email protected]>

commit 62a8ff67eba52dae9b107e1fb8827054ed00a265 upstream.

In certain scenarios a normal MSDU can be received as an A-MSDU when the A-MSDU present bit of a QoS header gets flipped during reception. Since this bit is unauthenticated, the hardware crypto engine can pass the frame to the driver without any error indication.

This could result in processing unintended subframes collected in the A-MSDU list. Hence, validate A-MSDU list by checking if the first frame has a valid subframe header.

Comparing the non-aggregated MSDU and an A-MSDU, the fields of the first subframe DA matches the LLC/SNAP header fields of a normal MSDU. In order to avoid processing such frames, add a validation to filter such A-MSDU frames where the first subframe header DA matches with the LLC/SNAP header pattern.

Tested-on: QCA9984 hw1.0 PCI 10.4-3.10-00047

Signed-off-by Sriram R <[email protected]>

Link https://lore.kernel.org/r/20210511200110.e6f5eb7b9847.I38a77ae26096862527a5eab73caebd7346af8b66@changeid open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/wireless/ath/ath10k/htt_rx.c |   61 ++++++++++++++++++++++++++++---
 1 file changed, 57 insertions(+), 4 deletions(-)

drivers/net/wireless/ath/ath10k/htt_rx.c +57 -4

@@ -2101,14 +2101,62 @@ static void ath10k_htt_rx_h_unchain(stru
 	ath10k_unchain_msdu(amsdu, unchain_cnt);
 }
 
+static bool ath10k_htt_rx_validate_amsdu(struct ath10k *ar,
+					 struct sk_buff_head *amsdu)
+{
+	u8 *subframe_hdr;
+	struct sk_buff *first;
+	bool is_first, is_last;
+	struct htt_rx_desc *rxd;
+	struct ieee80211_hdr *hdr;
+	size_t hdr_len, crypto_len;
+	enum htt_rx_mpdu_encrypt_type enctype;
+	int bytes_aligned = ar->hw_params.decap_align_bytes;
+
+	first = skb_peek(amsdu);
+
+	rxd = (void *)first->data - sizeof(*rxd);
+	hdr = (void *)rxd->rx_hdr_status;
+
+	is_first = !!(rxd->msdu_end.common.info0 &
+		      __cpu_to_le32(RX_MSDU_END_INFO0_FIRST_MSDU));
+	is_last = !!(rxd->msdu_end.common.info0 &
+		     __cpu_to_le32(RX_MSDU_END_INFO0_LAST_MSDU));
+
+	/* Return in case of non-aggregated msdu */
+	if (is_first && is_last)
+		return true;
+
+	/* First msdu flag is not set for the first msdu of the list */
+	if (!is_first)
+		return false;
+
+	enctype = MS(__le32_to_cpu(rxd->mpdu_start.info0),
+		     RX_MPDU_START_INFO0_ENCRYPT_TYPE);
+
+	hdr_len = ieee80211_hdrlen(hdr->frame_control);
+	crypto_len = ath10k_htt_rx_crypto_param_len(ar, enctype);
+
+	subframe_hdr = (u8 *)hdr + round_up(hdr_len, bytes_aligned) +
+		       crypto_len;
+
+	/* Validate if the amsdu has a proper first subframe.
+	 * There are chances a single msdu can be received as amsdu when
+	 * the unauthenticated amsdu flag of a QoS header
+	 * gets flipped in non-SPP AMSDU's, in such cases the first
+	 * subframe has llc/snap header in place of a valid da.
+	 * return false if the da matches rfc1042 pattern
+	 */
+	if (ether_addr_equal(subframe_hdr, rfc1042_header))
+		return false;
+
+	return true;
+}
+
 static bool ath10k_htt_rx_amsdu_allowed(struct ath10k *ar,
 					struct sk_buff_head *amsdu,
 					struct ieee80211_rx_status *rx_status)
 {
-	/* FIXME: It might be a good idea to do some fuzzy-testing to drop
-	 * invalid/dangerous frames.
-	 */
-
 	if (!rx_status->freq) {
 		ath10k_dbg(ar, ATH10K_DBG_HTT, "no channel configured; ignoring frame(s)!\n");
 		return false;

@@ -2119,6 +2167,11 @@ static bool ath10k_htt_rx_amsdu_allowed(
 		return false;
 	}
 
+	if (!ath10k_htt_rx_validate_amsdu(ar, amsdu)) {
+		ath10k_dbg(ar, ATH10K_DBG_HTT, "invalid amsdu received\n");
+		return false;
+	}
+
 	return true;
 }

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #30 lore raw

PATCH 30/177 dm snapshot: properly fix a crash when an origin has no snapshots

From: Mikulas Patocka <[email protected]>

commit 7e768532b2396bcb7fbf6f82384b85c0f1d2f197 upstream.

If an origin target has no snapshots, o->split_boundary is set to 0. This causes BUG_ON(sectors <= 0) in block/bio.c:bio_split().

Fix this by initializing chunk_size, and in turn split_boundary, to rounddown_pow_of_two(UINT_MAX) -- the largest power of two that fits into "unsigned" type.

Signed-off-by Mikulas Patocka <[email protected]>

Signed-off-by Mike Snitzer <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/md/dm-snap.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/md/dm-snap.c +1 -1

@@ -854,7 +854,7 @@ static int dm_add_exception(void *contex
 static uint32_t __minimum_chunk_size(struct origin *o)
 {
 	struct dm_snapshot *snap;
-	unsigned chunk_size = 0;
+	unsigned chunk_size = rounddown_pow_of_two(UINT_MAX);
 
 	if (o)
 		list_for_each_entry(snap, &o->snapshots, list)

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #31 lore raw

PATCH 31/177 drm/amdgpu/vcn1: add cancel_delayed_work_sync before power gate

From: James Zhu <[email protected]>

commit b95f045ea35673572ef46d6483ad8bd6d353d63c upstream.

Add cancel_delayed_work_sync before set power gating state to avoid race condition issue when power gating.

Signed-off-by James Zhu <[email protected]>

Reviewed-by Leo Liu <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +5 -1

@@ -233,9 +233,13 @@ static int vcn_v1_0_hw_fini(void *handle
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec;
 
+	cancel_delayed_work_sync(&adev->vcn.idle_work);
+
 	if ((adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) ||
-		RREG32_SOC15(VCN, 0, mmUVD_STATUS))
+		(adev->vcn.cur_state != AMD_PG_STATE_GATE &&
+		 RREG32_SOC15(VCN, 0, mmUVD_STATUS))) {
 		vcn_v1_0_set_powergating_state(adev, AMD_PG_STATE_GATE);
+	}
 
 	ring->sched.ready = false;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #32 lore raw

PATCH 32/177 drm/amdgpu/vcn2.0: add cancel_delayed_work_sync before power gate

From: James Zhu <[email protected]>

commit 0c6013377b4027e69d8f3e63b6bf556b6cb87802 upstream.

Add cancel_delayed_work_sync before set power gating state to avoid race condition issue when power gating.

Signed-off-by James Zhu <[email protected]>

Reviewed-by Leo Liu <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c |    2 ++
 1 file changed, 2 insertions(+)

drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +2 -0

@@ -293,6 +293,8 @@ static int vcn_v2_0_hw_fini(void *handle
 	struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec;
 	int i;
 
+	cancel_delayed_work_sync(&adev->vcn.idle_work);
+
 	if ((adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) ||
 	    (adev->vcn.cur_state != AMD_PG_STATE_GATE &&
 	      RREG32_SOC15(VCN, 0, mmUVD_STATUS)))

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #33 lore raw

PATCH 33/177 drm/amdgpu/vcn2.5: add cancel_delayed_work_sync before power gate

From: James Zhu <[email protected]>

commit 2fb536ea42d557f39f70c755f68e1aa1ad466c55 upstream.

Add cancel_delayed_work_sync before set power gating state to avoid race condition issue when power gating.

Signed-off-by James Zhu <[email protected]>

Reviewed-by Leo Liu <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c |    2 ++
 1 file changed, 2 insertions(+)

drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +2 -0

@@ -302,6 +302,8 @@ static int vcn_v2_5_hw_fini(void *handle
 	struct amdgpu_ring *ring;
 	int i;
 
+	cancel_delayed_work_sync(&adev->vcn.idle_work);
+
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #34 lore raw

PATCH 34/177 selftests/gpio: Use TEST_GEN_PROGS_EXTENDED

From: Michael Ellerman <[email protected]>

[ Upstream commit ff2c395b9257f0e617f9cd212893f3c72c80ee6c ]

Use TEST_GEN_PROGS_EXTENDED rather than TEST_PROGS_EXTENDED.

That tells the lib.mk logic that the files it references are to be generated by the Makefile.

Having done that we don't need to override the all rule.

Signed-off-by Michael Ellerman <[email protected]>

Signed-off-by Shuah Khan <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 tools/testing/selftests/gpio/Makefile | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

tools/testing/selftests/gpio/Makefile +3 -5

index 0bb80619db58..1963beaf5119 100644
@@ -11,22 +11,20 @@ LDLIBS += $(MOUNT_LDLIBS)
 
 TEST_PROGS := gpio-mockup.sh
 TEST_FILES := gpio-mockup-sysfs.sh
-TEST_PROGS_EXTENDED := gpio-mockup-chardev
+TEST_GEN_PROGS_EXTENDED := gpio-mockup-chardev
 
 GPIODIR := $(realpath ../../../gpio)
 GPIOOBJ := gpio-utils.o
 
-all: $(TEST_PROGS_EXTENDED)
-
 override define CLEAN
-	$(RM) $(TEST_PROGS_EXTENDED)
+	$(RM) $(TEST_GEN_PROGS_EXTENDED)
 	$(MAKE) -C $(GPIODIR) OUTPUT=$(GPIODIR)/ clean
 endef
 
 KSFT_KHDR_INSTALL := 1
 include ../lib.mk
 
-$(TEST_PROGS_EXTENDED): $(GPIODIR)/$(GPIOOBJ)
+$(TEST_GEN_PROGS_EXTENDED): $(GPIODIR)/$(GPIOOBJ)
 
 $(GPIODIR)/$(GPIOOBJ):
 	$(MAKE) OUTPUT=$(GPIODIR)/ -C $(GPIODIR)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #35 lore raw

PATCH 35/177 selftests/gpio: Move include of lib.mk up

From: Michael Ellerman <[email protected]>

[ Upstream commit 449539da2e237336bc750b41f1736a77f9aca25c ]

Move the include of lib.mk up so that in a subsequent patch we can use OUTPUT, which is initialised by lib.mk, in the definition of the GPIO variables.

Signed-off-by Michael Ellerman <[email protected]>

Signed-off-by Shuah Khan <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 tools/testing/selftests/gpio/Makefile | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

tools/testing/selftests/gpio/Makefile +3 -3

index 1963beaf5119..0b739bb4f2ad 100644
@@ -13,6 +13,9 @@ TEST_PROGS := gpio-mockup.sh
 TEST_FILES := gpio-mockup-sysfs.sh
 TEST_GEN_PROGS_EXTENDED := gpio-mockup-chardev
 
+KSFT_KHDR_INSTALL := 1
+include ../lib.mk
+
 GPIODIR := $(realpath ../../../gpio)
 GPIOOBJ := gpio-utils.o

@@ -21,9 +24,6 @@ override define CLEAN
 	$(MAKE) -C $(GPIODIR) OUTPUT=$(GPIODIR)/ clean
 endef
 
-KSFT_KHDR_INSTALL := 1
-include ../lib.mk
-
 $(TEST_GEN_PROGS_EXTENDED): $(GPIODIR)/$(GPIOOBJ)
 
 $(GPIODIR)/$(GPIOOBJ):

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #36 lore raw

PATCH 36/177 selftests/gpio: Fix build when source tree is read only

From: Michael Ellerman <[email protected]>

[ Upstream commit b68c1c65dec5fb5186ebd33ce52059b4c6db8500 ]

Currently the gpio selftests fail to build if the source tree is read only:

make -j 160 -C tools/testing/selftests TARGETS=gpio
make[1]: Entering directory '/linux/tools/testing/selftests/gpio'
make OUTPUT=/linux/tools/gpio/ -C /linux/tools/gpio
make[2]: Entering directory '/linux/tools/gpio'
mkdir -p /linux/tools/gpio/include/linux 2>&1 || true
ln -sf /linux/tools/gpio/../../include/uapi/linux/gpio.h /linux/tools/gpio/include/linux/gpio.h
ln: failed to create symbolic link '/linux/tools/gpio/include/linux/gpio.h': Read-only file system

This happens because we ask make to build ../../../gpio (tools/gpio) without pointing OUTPUT away from the source directory.

To fix it we create a subdirectory of the existing OUTPUT directory, called tools-gpio, and tell tools/gpio to build in there.

Signed-off-by Michael Ellerman <[email protected]>

Signed-off-by Shuah Khan <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 tools/testing/selftests/gpio/Makefile | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

tools/testing/selftests/gpio/Makefile +9 -5

index 0b739bb4f2ad..f270b6abd64c 100644
@@ -17,14 +17,18 @@ KSFT_KHDR_INSTALL := 1
 include ../lib.mk
 
 GPIODIR := $(realpath ../../../gpio)
-GPIOOBJ := gpio-utils.o
+GPIOOUT := $(OUTPUT)/tools-gpio/
+GPIOOBJ := $(GPIOOUT)/gpio-utils.o
 
 override define CLEAN
 	$(RM) $(TEST_GEN_PROGS_EXTENDED)
-	$(MAKE) -C $(GPIODIR) OUTPUT=$(GPIODIR)/ clean
+	$(RM) -rf $(GPIOOUT)
 endef
 
-$(TEST_GEN_PROGS_EXTENDED): $(GPIODIR)/$(GPIOOBJ)
+$(TEST_GEN_PROGS_EXTENDED): $(GPIOOBJ)
 
-$(GPIODIR)/$(GPIOOBJ):
-	$(MAKE) OUTPUT=$(GPIODIR)/ -C $(GPIODIR)
+$(GPIOOUT):
+	mkdir -p $@
+
+$(GPIOOBJ): $(GPIOOUT)
+	$(MAKE) OUTPUT=$(GPIOOUT) -C $(GPIODIR)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:15 UTC #37 lore raw

PATCH 37/177 kgdb: fix gcc-11 warnings harder

From: Greg Kroah-Hartman <[email protected]>

commit bda7d3ab06f19c02dcef61fefcb9dd954dfd5e4f upstream.

40cc3a80bb42 ("kgdb: fix gcc-11 warning on indentation") tried to fix up the gcc-11 complaints in this file by just reformatting the #defines. That worked for gcc 11.1.0, but in gcc 11.1.1 as shipped by Fedora 34, the warning came back for one of the #defines.

Fix this up again by putting { } around the if statement, now it is quiet again.

Fixes 40cc3a80bb42 ("kgdb: fix gcc-11 warning on indentation")

Cc Arnd Bergmann <[email protected]>

Cc Daniel Thompson <[email protected]>

Cc Jason Wessel <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/misc/kgdbts.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/misc/kgdbts.c +2 -1

@@ -100,8 +100,9 @@
 		printk(KERN_INFO a);	\
 } while (0)
 #define v2printk(a...) do {		\
-	if (verbose > 1)		\
+	if (verbose > 1) {		\
 		printk(KERN_INFO a);	\
+	}				\
 	touch_nmi_watchdog();		\
 } while (0)
 #define eprintk(a...) do {		\

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #38 lore raw

PATCH 38/177 Documentation: seccomp: Fix user notification documentation

From: Sargun Dhillon <[email protected]>

commit aac902925ea646e461c95edc98a8a57eb0def917 upstream.

The documentation had some previously incorrect information about how userspace notifications (and responses) were handled due to a change from a previously proposed patchset.

Signed-off-by Sargun Dhillon <[email protected]>

Acked-by Tycho Andersen <[email protected]>

Acked-by Christian Brauner <[email protected]>

Signed-off-by Kees Cook <[email protected]>

Fixes 6a21cc50f0c7 ("seccomp: add a return code to trap to userspace")

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 Documentation/userspace-api/seccomp_filter.rst |   16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

Documentation/userspace-api/seccomp_filter.rst +8 -8

@@ -250,14 +250,14 @@ Users can read via ``ioctl(SECCOMP_IOCTL
 seccomp notification fd to receive a ``struct seccomp_notif``, which contains
 five members: the input length of the structure, a unique-per-filter ``id``,
 the ``pid`` of the task which triggered this request (which may be 0 if the
-task is in a pid ns not visible from the listener's pid namespace), a ``flags``
-member which for now only has ``SECCOMP_NOTIF_FLAG_SIGNALED``, representing
-whether or not the notification is a result of a non-fatal signal, and the
-``data`` passed to seccomp. Userspace can then make a decision based on this
-information about what to do, and ``ioctl(SECCOMP_IOCTL_NOTIF_SEND)`` a
-response, indicating what should be returned to userspace. The ``id`` member of
-``struct seccomp_notif_resp`` should be the same ``id`` as in ``struct
-seccomp_notif``.
+task is in a pid ns not visible from the listener's pid namespace). The
+notification also contains the ``data`` passed to seccomp, and a filters flag.
+The structure should be zeroed out prior to calling the ioctl.
+
+Userspace can then make a decision based on this information about what to do,
+and ``ioctl(SECCOMP_IOCTL_NOTIF_SEND)`` a response, indicating what should be
+returned to userspace. The ``id`` member of ``struct seccomp_notif_resp`` should
+be the same ``id`` as in ``struct seccomp_notif``.
 
 It is worth noting that ``struct seccomp_data`` contains the values of register
 arguments to the syscall, but does not contain pointers to memory. The task's

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #39 lore raw

PATCH 39/177 serial: core: fix suspicious security_locked_down() call

From: Ondrej Mosnacek <[email protected]>

commit 5e722b217ad3cf41f5504db80a68062df82b5242 upstream.

The commit that added this check did so in a very strange way - first security_locked_down() is called, its value stored into retval, and if it's nonzero, then an additional check is made for (change_irq || change_port), and if this is true, the function returns. However, if the goto exit branch is not taken, the code keeps the retval value and continues executing the function. Then, depending on whether uport->ops->verify_port is set, the retval value may or may not be reset to zero and eventually the error value from security_locked_down() may abort the function a few lines below.

I will go out on a limb and assume that this isn't the intended behavior and that an error value from security_locked_down() was supposed to abort the function only in case (change_irq || change_port) is true.

Note that security_locked_down() should be called last in any series of checks, since the SELinux implementation of this hook will do a check against the policy and generate an audit record in case of denial. If the operation was to carry on after calling security_locked_down(), then the SELinux denial record would be bogus.

See commit 59438b46471a ("security,lockdown,selinux: implement SELinux lockdown") for how SELinux implements this hook.

Fixes 794edf30ee6c ("lockdown: Lock down TIOCSSERIAL")

Acked-by Kees Cook <[email protected]>

Signed-off-by Ondrej Mosnacek <[email protected]>

Cc stable <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/tty/serial/serial_core.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

drivers/tty/serial/serial_core.c +5 -3

@@ -863,9 +863,11 @@ static int uart_set_info(struct tty_stru
 		goto check_and_exit;
 	}
 
-	retval = security_locked_down(LOCKDOWN_TIOCSSERIAL);
-	if (retval && (change_irq || change_port))
-		goto exit;
+	if (change_irq || change_port) {
+		retval = security_locked_down(LOCKDOWN_TIOCSSERIAL);
+		if (retval)
+			goto exit;
+	}
 
 	/*
 	 * Ask the low level driver to verify the settings.

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #40 lore raw

PATCH 40/177 misc/uss720: fix memory leak in uss720_probe

From: Dongliang Mu <[email protected]>

commit dcb4b8ad6a448532d8b681b5d1a7036210b622de upstream.

uss720_probe forgets to decrease the refcount of usbdev in uss720_probe. Fix this by decreasing the refcount of usbdev by usb_put_dev.

BUG: memory leak
unreferenced object 0xffff888101113800 (size 2048):
  comm "kworker/0:1", pid 7, jiffies 4294956777 (age 28.870s)
  hex dump (first 32 bytes):
    ff ff ff ff 31 00 00 00 00 00 00 00 00 00 00 00  ....1...........
    00 00 00 00 00 00 00 00 00 00 00 00 03 00 00 00  ................
  backtrace:
    [<ffffffff82b8e822>] kmalloc include/linux/slab.h:554 [inline]
    [<ffffffff82b8e822>] kzalloc include/linux/slab.h:684 [inline]
    [<ffffffff82b8e822>] usb_alloc_dev+0x32/0x450 drivers/usb/core/usb.c:582
    [<ffffffff82b98441>] hub_port_connect drivers/usb/core/hub.c:5129 [inline]
    [<ffffffff82b98441>] hub_port_connect_change drivers/usb/core/hub.c:5363 [inline]
    [<ffffffff82b98441>] port_event drivers/usb/core/hub.c:5509 [inline]
    [<ffffffff82b98441>] hub_event+0x1171/0x20c0 drivers/usb/core/hub.c:5591
    [<ffffffff81259229>] process_one_work+0x2c9/0x600 kernel/workqueue.c:2275
    [<ffffffff81259b19>] worker_thread+0x59/0x5d0 kernel/workqueue.c:2421
    [<ffffffff81261228>] kthread+0x178/0x1b0 kernel/kthread.c:292
    [<ffffffff8100227f>] ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294

Fixes 0f36163d3abe ("[PATCH] usb: fix uss720 schedule with interrupts off")

Cc stable <[email protected]>

Reported-by [email protected]

Signed-off-by Dongliang Mu <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/misc/uss720.c |    1 +
 1 file changed, 1 insertion(+)

drivers/usb/misc/uss720.c +1 -0

@@ -736,6 +736,7 @@ static int uss720_probe(struct usb_inter
 	parport_announce_port(pp);
 
 	usb_set_intfdata(intf, pp);
+	usb_put_dev(usbdev);
 	return 0;
 
 probe_abort:

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #41 lore raw

PATCH 41/177 thunderbolt: dma_port: Fix NVM read buffer bounds and offset issue

From: Mathias Nyman <[email protected]>

commit b106776080a1cf953a1b2fd50cb2a995db4732be upstream.

Up to 64 bytes of data can be read from NVM in one go. Read address must be dword aligned. Data is read into a local buffer.

If caller asks to read data starting at an unaligned address then full dword is anyway read from NVM into a local buffer. Data is then copied from the local buffer starting at the unaligned offset to the caller buffer.

In cases where asked data length + unaligned offset is over 64 bytes we need to make sure we don't read past the 64 bytes in the local buffer when copying to caller buffer, and make sure that we don't skip copying unaligned offset bytes from local buffer anymore after the first round of 64 byte NVM data read.

Fixes 3e13676862f9 ("thunderbolt: Add support for DMA configuration based mailbox")

Signed-off-by Mathias Nyman <[email protected]>

Signed-off-by Mika Westerberg <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/thunderbolt/dma_port.c |   11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

drivers/thunderbolt/dma_port.c +6 -5

@@ -364,15 +364,15 @@ int dma_port_flash_read(struct tb_dma_po
 			void *buf, size_t size)
 {
 	unsigned int retries = DMA_PORT_RETRIES;
-	unsigned int offset;
-
-	offset = address & 3;
-	address = address & ~3;
 
 	do {
-		u32 nbytes = min_t(u32, size, MAIL_DATA_DWORDS * 4);
+		unsigned int offset;
+		size_t nbytes;
 		int ret;
 
+		offset = address & 3;
+		nbytes = min_t(size_t, size + offset, MAIL_DATA_DWORDS * 4);
+
 		ret = dma_port_flash_read_block(dma, address, dma->buf,
 						ALIGN(nbytes, 4));
 		if (ret) {

@@ -384,6 +384,7 @@ int dma_port_flash_read(struct tb_dma_po
 			return ret;
 		}
 
+		nbytes -= offset;
 		memcpy(buf, dma->buf + offset, nbytes);
 
 		size -= nbytes;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #42 lore raw

PATCH 42/177 mei: request autosuspend after sending rx flow control

From: Alexander Usyskin <[email protected]>

commit bbf0a94744edfeee298e4a9ab6fd694d639a5cdf upstream.

A rx flow control waiting in the control queue may block autosuspend. Re-request autosuspend after flow control been sent to unblock the transition to the low power state.

Signed-off-by Alexander Usyskin <[email protected]>

Signed-off-by Tomas Winkler <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/misc/mei/interrupt.c |    3 +++
 1 file changed, 3 insertions(+)

drivers/misc/mei/interrupt.c +3 -0

@@ -222,6 +222,9 @@ static int mei_cl_irq_read(struct mei_cl
 		return ret;
 	}
 
+	pm_runtime_mark_last_busy(dev->dev);
+	pm_request_autosuspend(dev->dev);
+
 	list_move_tail(&cb->list, &cl->rd_pending);
 
 	return 0;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #43 lore raw

PATCH 43/177 staging: iio: cdc: ad7746: avoid overwrite of num_channels

From: Lucas Stankus <[email protected]>

commit 04f5b9f539ce314f758d919a14dc7a669f3b7838 upstream.

AD7745 devices don't have the CIN2 pins and therefore can't handle related channels. Forcing the number of AD7746 channels may lead to enabling more channels than what the hardware actually supports.
Avoid num_channels being overwritten after first assignment.

Signed-off-by Lucas Stankus <[email protected]>

Fixes 83e416f458d53 ("staging: iio: adc: Replace, rewrite ad7745 from scratch.")

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/staging/iio/cdc/ad7746.c |    1 -
 1 file changed, 1 deletion(-)

drivers/staging/iio/cdc/ad7746.c +0 -1

@@ -702,7 +702,6 @@ static int ad7746_probe(struct i2c_clien
 		indio_dev->num_channels = ARRAY_SIZE(ad7746_channels);
 	else
 		indio_dev->num_channels =  ARRAY_SIZE(ad7746_channels) - 2;
-	indio_dev->num_channels = ARRAY_SIZE(ad7746_channels);
 	indio_dev->modes = INDIO_DIRECT_MODE;
 
 	if (pdata) {

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #44 lore raw

PATCH 44/177 iio: gyro: fxas21002c: balance runtime power in error path

Link https://lore.kernel.org/linux-iio/CBBZA9T1OY9C.2611WSV49DV2G@arch-thunder/ open here

From: Rui Miguel Silva <[email protected]>

commit 2a54c8c9ebc2006bf72554afc84ffc67768979a0 upstream.

If we fail to read temperature or axis we need to decrement the runtime pm reference count to trigger autosuspend.

Add the call to pm_put to do that in case of error.

Fixes a0701b6263ae ("iio: gyro: add core driver for fxas21002c")

Signed-off-by Rui Miguel Silva <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/iio/gyro/fxas21002c_core.c |    2 ++
 1 file changed, 2 insertions(+)

drivers/iio/gyro/fxas21002c_core.c +2 -0

@@ -333,6 +333,7 @@ static int fxas21002c_temp_get(struct fx
 	ret = regmap_field_read(data->regmap_fields[F_TEMP], &temp);
 	if (ret < 0) {
 		dev_err(dev, "failed to read temp: %d\n", ret);
+		fxas21002c_pm_put(data);
 		goto data_unlock;
 	}

@@ -366,6 +367,7 @@ static int fxas21002c_axis_get(struct fx
 			       &axis_be, sizeof(axis_be));
 	if (ret < 0) {
 		dev_err(dev, "failed to read axis: %d: %d\n", index, ret);
+		fxas21002c_pm_put(data);
 		goto data_unlock;
 	}

Greg Kroah-Hartman

author maintainer

2021-05-31 14:16 UTC #45 lore raw

PATCH 45/177 iio: adc: ad7768-1: Fix too small buffer passed to iio_push_to_buffers_with_timestamp()

From: Jonathan Cameron <[email protected]>

commit a1caeebab07e9d72eec534489f47964782b93ba9 upstream.

Add space for the timestamp to be inserted. Also ensure correct alignment for passing to iio_push_to_buffers_with_timestamp()

Fixes a5f8c7da3dbe ("iio: adc: Add AD7768-1 ADC basic support")

Reviewed-by Andy Shevchenko <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/iio/adc/ad7768-1.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

drivers/iio/adc/ad7768-1.c +6 -2

@@ -166,6 +166,10 @@ struct ad7768_state {
 	 * transfer buffers to live in their own cache lines.
 	 */
 	union {
+		struct {
+			__be32 chan;
+			s64 timestamp;
+		} scan;
 		__be32 d32;
 		u8 d8[2];
 	} data ____cacheline_aligned;

@@ -459,11 +463,11 @@ static irqreturn_t ad7768_trigger_handle
 
 	mutex_lock(&st->lock);
 
-	ret = spi_read(st->spi, &st->data.d32, 3);
+	ret = spi_read(st->spi, &st->data.scan.chan, 3);
 	if (ret < 0)
 		goto err_unlock;
 
-	iio_push_to_buffers_with_timestamp(indio_dev, &st->data.d32,
+	iio_push_to_buffers_with_timestamp(indio_dev, &st->data.scan,
 					   iio_get_time_ns(indio_dev));
 
 	iio_trigger_notify_done(indio_dev->trig);

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #46 lore raw

PATCH 46/177 iio: adc: ad7124: Fix missbalanced regulator enable / disable on error.

From: Jonathan Cameron <[email protected]>

commit 4573472315f0fa461330545ff2aa2f6da0b1ae76 upstream.

If the devm_regulator_get() call succeeded but not the regulator_enable() then regulator_disable() would be called on a regulator that was not enabled.

Fix this by moving regulator enabling / disabling over to devm_ management via devm_add_action_or_reset.

Alexandru's sign-off here because he pulled Jonathan's patch into a larger set which Jonathan then applied.

Fixes b3af341bbd96 ("iio: adc: Add ad7124 support")

Reviewed-by Alexandru Ardelean <[email protected]>

Signed-off-by Alexandru Ardelean <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/iio/adc/ad7124.c |   29 +++++++++++++----------------
 1 file changed, 13 insertions(+), 16 deletions(-)

drivers/iio/adc/ad7124.c +13 ~~-16~~

@@ -564,6 +564,11 @@ static int ad7124_setup(struct ad7124_st
 	return ret;
 }
 
+static void ad7124_reg_disable(void *r)
+{
+	regulator_disable(r);
+}
+
 static int ad7124_probe(struct spi_device *spi)
 {
 	const struct spi_device_id *id;

@@ -607,17 +612,20 @@ static int ad7124_probe(struct spi_devic
 		ret = regulator_enable(st->vref[i]);
 		if (ret)
 			return ret;
+
+		ret = devm_add_action_or_reset(&spi->dev, ad7124_reg_disable,
+					       st->vref[i]);
+		if (ret)
+			return ret;
 	}
 
 	st->mclk = devm_clk_get(&spi->dev, "mclk");
-	if (IS_ERR(st->mclk)) {
-		ret = PTR_ERR(st->mclk);
-		goto error_regulator_disable;
-	}
+	if (IS_ERR(st->mclk))
+		return PTR_ERR(st->mclk);
 
 	ret = clk_prepare_enable(st->mclk);
 	if (ret < 0)
-		goto error_regulator_disable;
+		return ret;
 
 	ret = ad7124_soft_reset(st);
 	if (ret < 0)

@@ -643,11 +651,6 @@ error_remove_trigger:
 	ad_sd_cleanup_buffer_and_trigger(indio_dev);
 error_clk_disable_unprepare:
 	clk_disable_unprepare(st->mclk);
-error_regulator_disable:
-	for (i = ARRAY_SIZE(st->vref) - 1; i >= 0; i--) {
-		if (!IS_ERR_OR_NULL(st->vref[i]))
-			regulator_disable(st->vref[i]);
-	}
 
 	return ret;
 }

@@ -656,17 +659,11 @@ static int ad7124_remove(struct spi_devi
 {
 	struct iio_dev *indio_dev = spi_get_drvdata(spi);
 	struct ad7124_state *st = iio_priv(indio_dev);
-	int i;
 
 	iio_device_unregister(indio_dev);
 	ad_sd_cleanup_buffer_and_trigger(indio_dev);
 	clk_disable_unprepare(st->mclk);
 
-	for (i = ARRAY_SIZE(st->vref) - 1; i >= 0; i--) {
-		if (!IS_ERR_OR_NULL(st->vref[i]))
-			regulator_disable(st->vref[i]);
-	}
-
 	return 0;
 }

Greg Kroah-Hartman

author maintainer

2021-05-31 14:20 UTC #47 lore raw

PATCH 47/177 iio: adc: ad7124: Fix potential overflow due to non sequential channel numbers

From: Jonathan Cameron <[email protected]>

commit f2a772c51206b0c3f262e4f6a3812c89a650191b upstream.

Channel numbering must start at 0 and then not have any holes, or it is possible to overflow the available storage. Note this bug was introduced as part of a fix to ensure we didn't rely on the ordering of child nodes. So we need to support arbitrary ordering but they all need to be there somewhere.

Note I hit this when using qemu to test the rest of this series. Arguably this isn't the best fix, but it is probably the most minimal option for backporting etc.

Alexandru's sign-off is here because he carried this patch in a larger set that Jonathan then applied.

Fixes d7857e4ee1ba6 ("iio: adc: ad7124: Fix DT channel configuration")

Reviewed-by Alexandru Ardelean <[email protected]>

Signed-off-by Alexandru Ardelean <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/iio/adc/ad7124.c |    7 +++++++
 1 file changed, 7 insertions(+)

drivers/iio/adc/ad7124.c +7 -0

@@ -473,6 +473,13 @@ static int ad7124_of_parse_channel_confi
 		if (ret)
 			goto err;
 
+		if (channel >= indio_dev->num_channels) {
+			dev_err(indio_dev->dev.parent,
+				"Channel index >= number of channels\n");
+			ret = -EINVAL;
+			goto err;
+		}
+
 		ret = of_property_read_u32_array(child, "diff-channels",
 						 ain, 2);
 		if (ret)

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #48 lore raw

PATCH 48/177 iio: adc: ad7793: Add missing error code in ad7793_setup()

From: YueHaibing <[email protected]>

commit 4ed243b1da169bcbc1ec5507867e56250c5f1ff9 upstream.

Set error code while device ID query failed.

Fixes 88bc30548aae ("IIO: ADC: New driver for AD7792/AD7793 3 Channel SPI ADC")

Signed-off-by YueHaibing <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/iio/adc/ad7793.c |    1 +
 1 file changed, 1 insertion(+)

drivers/iio/adc/ad7793.c +1 -0

@@ -278,6 +278,7 @@ static int ad7793_setup(struct iio_dev *
 	id &= AD7793_ID_MASK;
 
 	if (id != st->chip_info->id) {
+		ret = -ENODEV;
 		dev_err(&st->sd.spi->dev, "device ID query failed\n");
 		goto out;
 	}

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #49 lore raw

PATCH 49/177 serial: 8250_pci: Add support for new HPE serial device

From: Randy Wright <[email protected]>

commit e0e24208792080135248f23fdf6d51aa2e04df05 upstream.

Add support for new HPE serial device. It is MSI enabled, but otherwise similar to legacy HP server serial devices.

Tested-by Jerry Hoemann <[email protected]>

Signed-off-by Randy Wright <[email protected]>

Cc stable <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/tty/serial/8250/8250_pci.c |   18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

drivers/tty/serial/8250/8250_pci.c +18 -0

@@ -58,6 +58,8 @@ struct serial_private {
 	int			line[0];
 };
 
+#define PCI_DEVICE_ID_HPE_PCI_SERIAL	0x37e
+
 static const struct pci_device_id pci_use_msi[] = {
 	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_NETMOS, PCI_DEVICE_ID_NETMOS_9900,
 			 0xA000, 0x1000) },

@@ -65,6 +67,8 @@ static const struct pci_device_id pci_us
 			 0xA000, 0x1000) },
 	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_NETMOS, PCI_DEVICE_ID_NETMOS_9922,
 			 0xA000, 0x1000) },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_HP_3PAR, PCI_DEVICE_ID_HPE_PCI_SERIAL,
+			 PCI_ANY_ID, PCI_ANY_ID) },
 	{ }
 };

@@ -1966,6 +1970,16 @@ static struct pci_serial_quirk pci_seria
 		.setup		= pci_hp_diva_setup,
 	},
 	/*
+	 * HPE PCI serial device
+	 */
+	{
+		.vendor         = PCI_VENDOR_ID_HP_3PAR,
+		.device         = PCI_DEVICE_ID_HPE_PCI_SERIAL,
+		.subvendor      = PCI_ANY_ID,
+		.subdevice      = PCI_ANY_ID,
+		.setup		= pci_hp_diva_setup,
+	},
+	/*
 	 * Intel
 	 */
 	{

@@ -4932,6 +4946,10 @@ static const struct pci_device_id serial
 	{	PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_DIVA_AUX,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_b2_1_115200 },
+	/* HPE PCI serial device */
+	{	PCI_VENDOR_ID_HP_3PAR, PCI_DEVICE_ID_HPE_PCI_SERIAL,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
+		pbn_b1_1_115200 },
 
 	{	PCI_VENDOR_ID_DCI, PCI_DEVICE_ID_DCI_PCCOM2,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #50 lore raw

PATCH 50/177 serial: 8250_pci: handle FL_NOIRQ board flag

From: Christian Gmeiner <[email protected]>

commit 9808f9be31c68af43f6e531f2c851ebb066513fe upstream.

In commit 8428413b1d14 ("serial: 8250_pci: Implement MSI(-X) support") the way the irq gets allocated was changed. With that change the handling FL_NOIRQ got lost. Restore the old behaviour.

Fixes 8428413b1d14 ("serial: 8250_pci: Implement MSI(-X) support")

Signed-off-by Christian Gmeiner <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/tty/serial/8250/8250_pci.c |   29 +++++++++++++++++------------
 1 file changed, 17 insertions(+), 12 deletions(-)

drivers/tty/serial/8250/8250_pci.c +17 ~~-12~~

@@ -3917,21 +3917,26 @@ pciserial_init_ports(struct pci_dev *dev
 	uart.port.flags = UPF_SKIP_TEST | UPF_BOOT_AUTOCONF | UPF_SHARE_IRQ;
 	uart.port.uartclk = board->base_baud * 16;
 
-	if (pci_match_id(pci_use_msi, dev)) {
-		dev_dbg(&dev->dev, "Using MSI(-X) interrupts\n");
-		pci_set_master(dev);
-		rc = pci_alloc_irq_vectors(dev, 1, 1, PCI_IRQ_ALL_TYPES);
+	if (board->flags & FL_NOIRQ) {
+		uart.port.irq = 0;
 	} else {
-		dev_dbg(&dev->dev, "Using legacy interrupts\n");
-		rc = pci_alloc_irq_vectors(dev, 1, 1, PCI_IRQ_LEGACY);
-	}
-	if (rc < 0) {
-		kfree(priv);
-		priv = ERR_PTR(rc);
-		goto err_deinit;
+		if (pci_match_id(pci_use_msi, dev)) {
+			dev_dbg(&dev->dev, "Using MSI(-X) interrupts\n");
+			pci_set_master(dev);
+			rc = pci_alloc_irq_vectors(dev, 1, 1, PCI_IRQ_ALL_TYPES);
+		} else {
+			dev_dbg(&dev->dev, "Using legacy interrupts\n");
+			rc = pci_alloc_irq_vectors(dev, 1, 1, PCI_IRQ_LEGACY);
+		}
+		if (rc < 0) {
+			kfree(priv);
+			priv = ERR_PTR(rc);
+			goto err_deinit;
+		}
+
+		uart.port.irq = pci_irq_vector(dev, 0);
 	}
 
-	uart.port.irq = pci_irq_vector(dev, 0);
 	uart.port.dev = &dev->dev;
 
 	for (i = 0; i < nr_ports; i++) {

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #51 lore raw

PATCH 51/177 USB: trancevibrator: fix control-request direction

From: Johan Hovold <[email protected]>

commit 746e4acf87bcacf1406e05ef24a0b7139147c63e upstream.

The direction of the pipe argument must match the request-type direction bit or control requests may fail depending on the host-controller-driver implementation.

Fix the set-speed request which erroneously used USB_DIR_IN and update the default timeout argument to match (same value).

Fixes 5638e4d92e77 ("USB: add PlayStation 2 Trance Vibrator driver")

Cc [email protected] # 2.6.19

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/misc/trancevibrator.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/usb/misc/trancevibrator.c +2 -2

@@ -61,9 +61,9 @@ static ssize_t speed_store(struct device
 	/* Set speed */
 	retval = usb_control_msg(tv->udev, usb_sndctrlpipe(tv->udev, 0),
 				 0x01, /* vendor request: set speed */
-				 USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_OTHER,
+				 USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_OTHER,
 				 tv->speed, /* speed value */
-				 0, NULL, 0, USB_CTRL_GET_TIMEOUT);
+				 0, NULL, 0, USB_CTRL_SET_TIMEOUT);
 	if (retval) {
 		tv->speed = old;
 		dev_dbg(&tv->udev->dev, "retval = %d\n", retval);

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #52 lore raw

PATCH 52/177 USB: usbfs: Dont WARN about excessively large memory allocations

From: Alan Stern <[email protected]>

commit 4f2629ea67e7225c3fd292c7fe4f5b3c9d6392de upstream.

Syzbot found that the kernel generates a WARNing if the user tries to submit a bulk transfer through usbfs with a buffer that is way too large. This isn't a bug in the kernel; it's merely an invalid request from the user and the usbfs code does handle it correctly.

In theory the same thing can happen with async transfers, or with the packet descriptor table for isochronous transfers.

To prevent the MM subsystem from complaining about these bad allocation requests, add the __GFP_NOWARN flag to the kmalloc calls for these buffers.

CC Andrew Morton <[email protected]>

CC <[email protected]>

Reported-and-tested-by [email protected]

Signed-off-by Alan Stern <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/core/devio.c |   11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

drivers/usb/core/devio.c +8 -3

@@ -1218,7 +1218,12 @@ static int proc_bulk(struct usb_dev_stat
 	ret = usbfs_increase_memory_usage(len1 + sizeof(struct urb));
 	if (ret)
 		return ret;
-	tbuf = kmalloc(len1, GFP_KERNEL);
+
+	/*
+	 * len1 can be almost arbitrarily large.  Don't WARN if it's
+	 * too big, just fail the request.
+	 */
+	tbuf = kmalloc(len1, GFP_KERNEL | __GFP_NOWARN);
 	if (!tbuf) {
 		ret = -ENOMEM;
 		goto done;

@@ -1691,7 +1696,7 @@ static int proc_do_submiturb(struct usb_
 	if (num_sgs) {
 		as->urb->sg = kmalloc_array(num_sgs,
 					    sizeof(struct scatterlist),
-					    GFP_KERNEL);
+					    GFP_KERNEL | __GFP_NOWARN);
 		if (!as->urb->sg) {
 			ret = -ENOMEM;
 			goto error;

@@ -1726,7 +1731,7 @@ static int proc_do_submiturb(struct usb_
 					(uurb_start - as->usbm->vm_start);
 		} else {
 			as->urb->transfer_buffer = kmalloc(uurb->buffer_length,
-					GFP_KERNEL);
+					GFP_KERNEL | __GFP_NOWARN);
 			if (!as->urb->transfer_buffer) {
 				ret = -ENOMEM;
 				goto error;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #53 lore raw

PATCH 53/177 serial: tegra: Fix a mask operation that is always true

From: Colin Ian King <[email protected]>

commit 3ddb4ce1e6e3bd112778ab93bbd9092f23a878ec upstream.

Currently the expression lsr | UART_LSR_TEMT is always true and this seems suspect. I believe the intent was to mask lsr with UART_LSR_TEMT to check that bit, so the expression should be using the & operator instead. Fix this.

Fixes b9c2470fb150 ("serial: tegra: flush the RX fifo on frame error")

Signed-off-by Colin Ian King <[email protected]>

Cc stable <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/tty/serial/serial-tegra.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/tty/serial/serial-tegra.c +1 -1

@@ -332,7 +332,7 @@ static void tegra_uart_fifo_reset(struct
 
 	do {
 		lsr = tegra_uart_read(tup, UART_LSR);
-		if ((lsr | UART_LSR_TEMT) && !(lsr & UART_LSR_DR))
+		if ((lsr & UART_LSR_TEMT) && !(lsr & UART_LSR_DR))
 			break;
 		udelay(1);
 	} while (--tmout);

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #54 lore raw

PATCH 54/177 serial: sh-sci: Fix off-by-one error in FIFO threshold register setting

Link https://lore.kernel.org/r/5eff320aef92ffb33d00e57979fd3603bbb4a70f.1620648218.git.geert+renesas@glider.be open here

From: Geert Uytterhoeven <[email protected]>

commit 2ea2e019c190ee3973ef7bcaf829d8762e56e635 upstream.

The Receive FIFO Data Count Trigger field (RTRG[6:0]) in the Receive FIFO Data Count Trigger Register (HSRTRGR) of HSCIF can only hold values ranging from 0-127. As the FIFO size is equal to 128 on HSCIF, the user can write an out-of-range value, touching reserved bits.

Fix this by limiting the trigger value to the FIFO size minus one. Reverse the order of the checks, to avoid rx_trig becoming zero if the FIFO size is one.

Note that this change has no impact on other SCIF variants, as their maximum supported trigger value is lower than the FIFO size anyway, and the code below takes care of enforcing these limits.

Fixes a380ed461f66d1b8 ("serial: sh-sci: implement FIFO threshold register setting")

Reported-by Linh Phung <[email protected]>

Reviewed-by Wolfram Sang <[email protected]>

Reviewed-by Ulrich Hecht <[email protected]>

Signed-off-by Geert Uytterhoeven <[email protected]>

Cc stable <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/tty/serial/sh-sci.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/tty/serial/sh-sci.c +2 -2

@@ -1026,10 +1026,10 @@ static int scif_set_rtrg(struct uart_por
 {
 	unsigned int bits;
 
+	if (rx_trig >= port->fifosize)
+		rx_trig = port->fifosize - 1;
 	if (rx_trig < 1)
 		rx_trig = 1;
-	if (rx_trig >= port->fifosize)
-		rx_trig = port->fifosize;
 
 	/* HSCIF can be set to an arbitrary level. */
 	if (sci_getreg(port, HSRTRGR)->size) {

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #55 lore raw

PATCH 55/177 serial: rp2: use request_firmware instead of request_firmware_nowait

From: Zheyu Ma <[email protected]>

commit 016002848c82eeb5d460489ce392d91fe18c475c upstream.

In 'rp2_probe', the driver registers 'rp2_uart_interrupt' then calls 'rp2_fw_cb' through 'request_firmware_nowait'. In 'rp2_fw_cb', if the firmware don't exists, function just return without initializing ports of 'rp2_card'. But now the interrupt handler function has been registered, and when an interrupt comes, 'rp2_uart_interrupt' may access those ports then causing NULL pointer dereference or other bugs.

Because the driver does some initialization work in 'rp2_fw_cb', in order to make the driver ready to handle interrupts, 'request_firmware' should be used instead of asynchronous 'request_firmware_nowait'.

This report reveals it:

INFO: trying to register non-static key.
the code is fine but needs lockdep annotation.
turning off the locking correctness validator.

CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.19.177-gdba4159c14ef-dirty #45
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.0-59-
gc9ba5276e321-prebuilt.qemu.org 04/01/2014
Call Trace:
 <IRQ>
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0xec/0x156 lib/dump_stack.c:118
 assign_lock_key kernel/locking/lockdep.c:727 [inline]
 register_lock_class+0x14e5/0x1ba0 kernel/locking/lockdep.c:753
 __lock_acquire+0x187/0x3750 kernel/locking/lockdep.c:3303
 lock_acquire+0x124/0x340 kernel/locking/lockdep.c:3907
 __raw_spin_lock include/linux/spinlock_api_smp.h:142 [inline]
 _raw_spin_lock+0x32/0x50 kernel/locking/spinlock.c:144
 spin_lock include/linux/spinlock.h:329 [inline]
 rp2_ch_interrupt drivers/tty/serial/rp2.c:466 [inline]
 rp2_asic_interrupt.isra.9+0x15d/0x990 drivers/tty/serial/rp2.c:493
 rp2_uart_interrupt+0x49/0xe0 drivers/tty/serial/rp2.c:504
 __handle_irq_event_percpu+0xfb/0x770 kernel/irq/handle.c:149
 handle_irq_event_percpu+0x79/0x150 kernel/irq/handle.c:189
 handle_irq_event+0xac/0x140 kernel/irq/handle.c:206
 handle_fasteoi_irq+0x232/0x5c0 kernel/irq/chip.c:725
 generic_handle_irq_desc include/linux/irqdesc.h:155 [inline]
 handle_irq+0x230/0x3a0 arch/x86/kernel/irq_64.c:87
 do_IRQ+0xa7/0x1e0 arch/x86/kernel/irq.c:247
 common_interrupt+0xf/0xf arch/x86/entry/entry_64.S:670
 </IRQ>
RIP: 0010:native_safe_halt+0x28/0x30 arch/x86/include/asm/irqflags.h:61
Code: 00 00 55 be 04 00 00 00 48 c7 c7 00 c2 2f 8c 48 89 e5 e8 fb 31 e7 f8

8b 05 75 af 8d 03 85 c0 7e 07 0f 00 2d 8a 61 65 00 fb f4 <5d> c3 90 90 90 90 90 90 0f 1f 44 00 00 55 48 89 e5 41 57 41 56 41
RSP: 0018:ffff88806b71fcc8 EFLAGS: 00000246 ORIG_RAX: ffffffffffffffde
RAX: 0000000000000000 RBX: ffffffff8bde7e48 RCX: ffffffff88a21285
RDX: 0000000000000000 RSI: 0000000000000004 RDI: ffffffff8c2fc200
RBP: ffff88806b71fcc8 R08: fffffbfff185f840 R09: fffffbfff185f840
R10: 0000000000000001 R11: fffffbfff185f840 R12: 0000000000000002
R13: ffffffff8bea18a0 R14: 0000000000000000 R15: 0000000000000000

 arch_safe_halt arch/x86/include/asm/paravirt.h:94 [inline]
 default_idle+0x6f/0x360 arch/x86/kernel/process.c:557
 arch_cpu_idle+0xf/0x20 arch/x86/kernel/process.c:548
 default_idle_call+0x3b/0x60 kernel/sched/idle.c:93
 cpuidle_idle_call kernel/sched/idle.c:153 [inline]
 do_idle+0x2ab/0x3c0 kernel/sched/idle.c:263
 cpu_startup_entry+0xcb/0xe0 kernel/sched/idle.c:369
 start_secondary+0x3b8/0x4e0 arch/x86/kernel/smpboot.c:271
 secondary_startup_64+0xa4/0xb0 arch/x86/kernel/head_64.S:243
BUG: unable to handle kernel NULL pointer dereference at 0000000000000010
PGD 8000000056d27067 P4D 8000000056d27067 PUD 56d28067 PMD 0
Oops: 0000 [#1] PREEMPT SMP KASAN PTI
CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.19.177-gdba4159c14ef-dirty #45
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.0-59-
gc9ba5276e321-prebuilt.qemu.org 04/01/2014
RIP: 0010:readl arch/x86/include/asm/io.h:59 [inline]
RIP: 0010:rp2_ch_interrupt drivers/tty/serial/rp2.c:472 [inline]
RIP: 0010:rp2_asic_interrupt.isra.9+0x181/0x990 drivers/tty/serial/rp2.c:
493
Code: df e8 43 5d c2 05 48 8d 83 e8 01 00 00 48 89 85 60 ff ff ff 48 c1 e8
03 42 80 3c 30 00 0f 85 aa 07 00 00 48 8b 83 e8 01 00 00 <8b> 40 10 89 c1
89 85 68 ff ff ff 48 8b 83 e8 01 00 00 89 48 10 83
RSP: 0018:ffff88806c287cd0 EFLAGS: 00010046
RAX: 0000000000000000 RBX: ffff88806ade6820 RCX: ffffffff814300b1
RDX: 1ffff1100d5bcd06 RSI: 0000000000000004 RDI: ffff88806ade6820
RBP: ffff88806c287db8 R08: ffffed100d5bcd05 R09: ffffed100d5bcd05
R10: 0000000000000001 R11: ffffed100d5bcd04 R12: ffffc90001e00000
R13: ffff888069654e10 R14: dffffc0000000000 R15: ffff888069654df0
FS:  0000000000000000(0000) GS:ffff88806c280000(0000) knlGS:
0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000010 CR3: 000000006892c000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <IRQ>
 rp2_uart_interrupt+0x49/0xe0 drivers/tty/serial/rp2.c:504
 __handle_irq_event_percpu+0xfb/0x770 kernel/irq/handle.c:149
 handle_irq_event_percpu+0x79/0x150 kernel/irq/handle.c:189
 handle_irq_event+0xac/0x140 kernel/irq/handle.c:206
 handle_fasteoi_irq+0x232/0x5c0 kernel/irq/chip.c:725
 generic_handle_irq_desc include/linux/irqdesc.h:155 [inline]
 handle_irq+0x230/0x3a0 arch/x86/kernel/irq_64.c:87
 do_IRQ+0xa7/0x1e0 arch/x86/kernel/irq.c:247
 common_interrupt+0xf/0xf arch/x86/entry/entry_64.S:670
 </IRQ>
RIP: 0010:native_safe_halt+0x28/0x30 arch/x86/include/asm/irqflags.h:61
Code: 00 00 55 be 04 00 00 00 48 c7 c7 00 c2 2f 8c 48 89 e5 e8 fb 31 e7

f8 8b 05 75 af 8d 03 85 c0 7e 07 0f 00 2d 8a 61 65 00 fb f4 <5d> c3 90 90 90 90 90 90 0f 1f 44 00 00 55 48 89 e5 41 57 41 56 41
RSP: 0018:ffff88806b71fcc8 EFLAGS: 00000246 ORIG_RAX: ffffffffffffffde
RAX: 0000000000000000 RBX: ffffffff8bde7e48 RCX: ffffffff88a21285
RDX: 0000000000000000 RSI: 0000000000000004 RDI: ffffffff8c2fc200
RBP: ffff88806b71fcc8 R08: fffffbfff185f840 R09: fffffbfff185f840
R10: 0000000000000001 R11: fffffbfff185f840 R12: 0000000000000002
R13: ffffffff8bea18a0 R14: 0000000000000000 R15: 0000000000000000

 arch_safe_halt arch/x86/include/asm/paravirt.h:94 [inline]
 default_idle+0x6f/0x360 arch/x86/kernel/process.c:557
 arch_cpu_idle+0xf/0x20 arch/x86/kernel/process.c:548
 default_idle_call+0x3b/0x60 kernel/sched/idle.c:93
 cpuidle_idle_call kernel/sched/idle.c:153 [inline]
 do_idle+0x2ab/0x3c0 kernel/sched/idle.c:263
 cpu_startup_entry+0xcb/0xe0 kernel/sched/idle.c:369
 start_secondary+0x3b8/0x4e0 arch/x86/kernel/smpboot.c:271
 secondary_startup_64+0xa4/0xb0 arch/x86/kernel/head_64.S:243
Modules linked in:
Dumping ftrace buffer:
   (ftrace buffer empty)
CR2: 0000000000000010
---[ end trace 11804dbb55cb1a64 ]---
RIP: 0010:readl arch/x86/include/asm/io.h:59 [inline]
RIP: 0010:rp2_ch_interrupt drivers/tty/serial/rp2.c:472 [inline]
RIP: 0010:rp2_asic_interrupt.isra.9+0x181/0x990 drivers/tty/serial/rp2.c:
493
Code: df e8 43 5d c2 05 48 8d 83 e8 01 00 00 48 89 85 60 ff ff ff 48 c1
e8 03 42 80 3c 30 00 0f 85 aa 07 00 00 48 8b 83 e8 01 00 00 <8b> 40 10 89
c1 89 85 68 ff ff ff 48 8b 83 e8 01 00 00 89 48 10 83
RSP: 0018:ffff88806c287cd0 EFLAGS: 00010046
RAX: 0000000000000000 RBX: ffff88806ade6820 RCX: ffffffff814300b1
RDX: 1ffff1100d5bcd06 RSI: 0000000000000004 RDI: ffff88806ade6820
RBP: ffff88806c287db8 R08: ffffed100d5bcd05 R09: ffffed100d5bcd05
R10: 0000000000000001 R11: ffffed100d5bcd04 R12: ffffc90001e00000
R13: ffff888069654e10 R14: dffffc0000000000 R15: ffff888069654df0
FS:  0000000000000000(0000) GS:ffff88806c280000(0000) knlGS:
0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000010 CR3: 000000006892c000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400

Reported-by Zheyu Ma <[email protected]>

Signed-off-by Zheyu Ma <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Cc stable <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/tty/serial/rp2.c |   52 +++++++++++++++--------------------------------
 1 file changed, 17 insertions(+), 35 deletions(-)

drivers/tty/serial/rp2.c +17 ~~-35~~

@@ -195,7 +195,6 @@ struct rp2_card {
 	void __iomem			*bar0;
 	void __iomem			*bar1;
 	spinlock_t			card_lock;
-	struct completion		fw_loaded;
 };
 
 #define RP_ID(prod) PCI_VDEVICE(RP, (prod))

@@ -664,17 +663,10 @@ static void rp2_remove_ports(struct rp2_
 	card->initialized_ports = 0;
 }
 
-static void rp2_fw_cb(const struct firmware *fw, void *context)
+static int rp2_load_firmware(struct rp2_card *card, const struct firmware *fw)
 {
-	struct rp2_card *card = context;
 	resource_size_t phys_base;
-	int i, rc = -ENOENT;
-
-	if (!fw) {
-		dev_err(&card->pdev->dev, "cannot find '%s' firmware image\n",
-			RP2_FW_NAME);
-		goto no_fw;
-	}
+	int i, rc = 0;
 
 	phys_base = pci_resource_start(card->pdev, 1);

@@ -720,23 +712,13 @@ static void rp2_fw_cb(const struct firmw
 		card->initialized_ports++;
 	}
 
-	release_firmware(fw);
-no_fw:
-	/*
-	 * rp2_fw_cb() is called from a workqueue long after rp2_probe()
-	 * has already returned success.  So if something failed here,
-	 * we'll just leave the now-dormant device in place until somebody
-	 * unbinds it.
-	 */
-	if (rc)
-		dev_warn(&card->pdev->dev, "driver initialization failed\n");
-
-	complete(&card->fw_loaded);
+	return rc;
 }
 
 static int rp2_probe(struct pci_dev *pdev,
 				   const struct pci_device_id *id)
 {
+	const struct firmware *fw;
 	struct rp2_card *card;
 	struct rp2_uart_port *ports;
 	void __iomem * const *bars;

@@ -747,7 +729,6 @@ static int rp2_probe(struct pci_dev *pde
 		return -ENOMEM;
 	pci_set_drvdata(pdev, card);
 	spin_lock_init(&card->card_lock);
-	init_completion(&card->fw_loaded);
 
 	rc = pcim_enable_device(pdev);
 	if (rc)

@@ -780,21 +761,23 @@ static int rp2_probe(struct pci_dev *pde
 		return -ENOMEM;
 	card->ports = ports;
 
-	rc = devm_request_irq(&pdev->dev, pdev->irq, rp2_uart_interrupt,
-			      IRQF_SHARED, DRV_NAME, card);
-	if (rc)
+	rc = request_firmware(&fw, RP2_FW_NAME, &pdev->dev);
+	if (rc < 0) {
+		dev_err(&pdev->dev, "cannot find '%s' firmware image\n",
+			RP2_FW_NAME);
 		return rc;
+	}
 
-	/*
-	 * Only catastrophic errors (e.g. ENOMEM) are reported here.
-	 * If the FW image is missing, we'll find out in rp2_fw_cb()
-	 * and print an error message.
-	 */
-	rc = request_firmware_nowait(THIS_MODULE, 1, RP2_FW_NAME, &pdev->dev,
-				     GFP_KERNEL, card, rp2_fw_cb);
+	rc = rp2_load_firmware(card, fw);
+
+	release_firmware(fw);
+	if (rc < 0)
+		return rc;
+
+	rc = devm_request_irq(&pdev->dev, pdev->irq, rp2_uart_interrupt,
+			      IRQF_SHARED, DRV_NAME, card);
 	if (rc)
 		return rc;
-	dev_dbg(&pdev->dev, "waiting for firmware blob...\n");
 
 	return 0;
 }

@@ -803,7 +786,6 @@ static void rp2_remove(struct pci_dev *p
 {
 	struct rp2_card *card = pci_get_drvdata(pdev);
 
-	wait_for_completion(&card->fw_loaded);
 	rp2_remove_ports(card);
 }

Greg Kroah-Hartman

author maintainer

2021-05-31 14:20 UTC #56 lore raw

PATCH 56/177 USB: serial: ti_usb_3410_5052: add startech.com device id

From: Sean MacLennan <[email protected]>

commit 89b1a3d811e6f8065d6ae8a25e7682329b4a31e2 upstream.

This adds support for the Startech.com generic serial to USB converter. It seems to be a bone stock TI_3410. I have been using this patch for years.

Signed-off-by Sean MacLennan <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/serial/ti_usb_3410_5052.c |    3 +++
 1 file changed, 3 insertions(+)

drivers/usb/serial/ti_usb_3410_5052.c +3 -0

@@ -37,6 +37,7 @@
 /* Vendor and product ids */
 #define TI_VENDOR_ID			0x0451
 #define IBM_VENDOR_ID			0x04b3
+#define STARTECH_VENDOR_ID		0x14b0
 #define TI_3410_PRODUCT_ID		0x3410
 #define IBM_4543_PRODUCT_ID		0x4543
 #define IBM_454B_PRODUCT_ID		0x454b

@@ -372,6 +373,7 @@ static const struct usb_device_id ti_id_
 	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1131_PRODUCT_ID) },
 	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1150_PRODUCT_ID) },
 	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1151_PRODUCT_ID) },
+	{ USB_DEVICE(STARTECH_VENDOR_ID, TI_3410_PRODUCT_ID) },
 	{ }	/* terminator */
 };

@@ -410,6 +412,7 @@ static const struct usb_device_id ti_id_
 	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1131_PRODUCT_ID) },
 	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1150_PRODUCT_ID) },
 	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1151_PRODUCT_ID) },
+	{ USB_DEVICE(STARTECH_VENDOR_ID, TI_3410_PRODUCT_ID) },
 	{ }	/* terminator */
 };

Greg Kroah-Hartman

author maintainer

2021-05-31 14:20 UTC #57 lore raw

PATCH 57/177 USB: serial: option: add Telit LE910-S1 compositions 0x7010, 0x7011

From: Daniele Palmas <[email protected]>

commit e467714f822b5d167a7fb03d34af91b5b6af1827 upstream.

Add support for the following Telit LE910-S1 compositions:

0x7010: rndis, tty, tty, tty
0x7011: ecm, tty, tty, tty

Signed-off-by Daniele Palmas <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/serial/option.c |    4 ++++
 1 file changed, 4 insertions(+)

drivers/usb/serial/option.c +4 -0

@@ -1240,6 +1240,10 @@ static const struct usb_device_id option
 	  .driver_info = NCTRL(0) | RSVD(1) },
 	{ USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x1901, 0xff),	/* Telit LN940 (MBIM) */
 	  .driver_info = NCTRL(0) },
+	{ USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x7010, 0xff),	/* Telit LE910-S1 (RNDIS) */
+	  .driver_info = NCTRL(2) },
+	{ USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, 0x7011, 0xff),	/* Telit LE910-S1 (ECM) */
+	  .driver_info = NCTRL(2) },
 	{ USB_DEVICE(TELIT_VENDOR_ID, 0x9010),				/* Telit SBL FN980 flashing device */
 	  .driver_info = NCTRL(0) | ZLP },
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, ZTE_PRODUCT_MF622, 0xff, 0xff, 0xff) }, /* ZTE WCDMA products */

Greg Kroah-Hartman

author maintainer

2021-05-31 14:20 UTC #58 lore raw

PATCH 58/177 USB: serial: ftdi_sio: add IDs for IDS GmbH Products

From: Dominik Andreas Schorpp <[email protected]>

commit c5a80540e425a5f9a82b0f3163e3b6a4331f33bc upstream.

Add the IDS GmbH Vendor ID and the Product IDs for SI31A (2xRS232) and CM31A (LoRaWAN Modem).

Signed-off-by Dominik Andreas Schorpp <[email protected]>

Signed-off-by Juergen Borleis <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/serial/ftdi_sio.c     |    3 +++
 drivers/usb/serial/ftdi_sio_ids.h |    7 +++++++
 2 files changed, 10 insertions(+)

drivers/usb/serial/ftdi_sio.c +3 -0

@@ -1034,6 +1034,9 @@ static const struct usb_device_id id_tab
 	/* Sienna devices */
 	{ USB_DEVICE(FTDI_VID, FTDI_SIENNA_PID) },
 	{ USB_DEVICE(ECHELON_VID, ECHELON_U20_PID) },
+	/* IDS GmbH devices */
+	{ USB_DEVICE(IDS_VID, IDS_SI31A_PID) },
+	{ USB_DEVICE(IDS_VID, IDS_CM31A_PID) },
 	/* U-Blox devices */
 	{ USB_DEVICE(UBLOX_VID, UBLOX_C099F9P_ZED_PID) },
 	{ USB_DEVICE(UBLOX_VID, UBLOX_C099F9P_ODIN_PID) },

drivers/usb/serial/ftdi_sio_ids.h +7 -0

@@ -1568,6 +1568,13 @@
 #define UNJO_ISODEBUG_V1_PID		0x150D
 
 /*
+ * IDS GmbH
+ */
+#define IDS_VID				0x2CAF
+#define IDS_SI31A_PID			0x13A2
+#define IDS_CM31A_PID			0x13A3
+
+/*
  * U-Blox products (http://www.u-blox.com).
  */
 #define UBLOX_VID			0x1546

Greg Kroah-Hartman

author maintainer

2021-05-31 14:20 UTC #59 lore raw

PATCH 59/177 USB: serial: pl2303: add device id for ADLINK ND-6530 GC

From: Zolton Jheng <[email protected]>

commit f8e8c1b2f782e7391e8a1c25648ce756e2a7d481 upstream.

This adds the device id for the ADLINK ND-6530 which is a PL2303GC based device.

Signed-off-by Zolton Jheng <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/serial/pl2303.c |    1 +
 drivers/usb/serial/pl2303.h |    1 +
 2 files changed, 2 insertions(+)

drivers/usb/serial/pl2303.c +1 -0

@@ -107,6 +107,7 @@ static const struct usb_device_id id_tab
 	{ USB_DEVICE(SONY_VENDOR_ID, SONY_QN3USB_PRODUCT_ID) },
 	{ USB_DEVICE(SANWA_VENDOR_ID, SANWA_PRODUCT_ID) },
 	{ USB_DEVICE(ADLINK_VENDOR_ID, ADLINK_ND6530_PRODUCT_ID) },
+	{ USB_DEVICE(ADLINK_VENDOR_ID, ADLINK_ND6530GC_PRODUCT_ID) },
 	{ USB_DEVICE(SMART_VENDOR_ID, SMART_PRODUCT_ID) },
 	{ USB_DEVICE(AT_VENDOR_ID, AT_VTKIT3_PRODUCT_ID) },
 	{ }					/* Terminating entry */

drivers/usb/serial/pl2303.h +1 -0

@@ -152,6 +152,7 @@
 /* ADLINK ND-6530 RS232,RS485 and RS422 adapter */
 #define ADLINK_VENDOR_ID		0x0b63
 #define ADLINK_ND6530_PRODUCT_ID	0x6530
+#define ADLINK_ND6530GC_PRODUCT_ID	0x653a
 
 /* SMART USB Serial Adapter */
 #define SMART_VENDOR_ID	0x0b8c

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #60 lore raw

PATCH 60/177 thermal/drivers/intel: Initialize RW trip to THERMAL_TEMP_INVALID

From: Srinivas Pandruvada <[email protected]>

commit eb8500b874cf295971a6a2a04e14eb0854197a3c upstream.

After commit 81ad4276b505 ("Thermal: Ignore invalid trip points") all user_space governor notifications via RW trip point is broken in intel thermal drivers. This commits marks trip_points with value of 0 during call to thermal_zone_device_register() as invalid. RW trip points can be 0 as user space will set the correct trip temperature later.

During driver init, x86_package_temp and all int340x drivers sets RW trip temperature as 0. This results in all these trips marked as invalid by the thermal core.

To fix this initialize RW trips to THERMAL_TEMP_INVALID instead of 0.

Signed-off-by Srinivas Pandruvada <[email protected]>

Signed-off-by Daniel Lezcano <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/thermal/intel/int340x_thermal/int340x_thermal_zone.c |    4 ++++
 drivers/thermal/intel/x86_pkg_temp_thermal.c                 |    2 +-
 2 files changed, 5 insertions(+), 1 deletion(-)

drivers/thermal/intel/int340x_thermal/int340x_thermal_zone.c +4 -0

@@ -230,6 +230,8 @@ struct int34x_thermal_zone *int340x_ther
 	if (ACPI_FAILURE(status))
 		trip_cnt = 0;
 	else {
+		int i;
+
 		int34x_thermal_zone->aux_trips =
 			kcalloc(trip_cnt,
 				sizeof(*int34x_thermal_zone->aux_trips),

@@ -240,6 +242,8 @@ struct int34x_thermal_zone *int340x_ther
 		}
 		trip_mask = BIT(trip_cnt) - 1;
 		int34x_thermal_zone->aux_trip_nr = trip_cnt;
+		for (i = 0; i < trip_cnt; ++i)
+			int34x_thermal_zone->aux_trips[i] = THERMAL_TEMP_INVALID;
 	}
 
 	trip_cnt = int340x_thermal_read_trips(int34x_thermal_zone);

drivers/thermal/intel/x86_pkg_temp_thermal.c +1 -1

@@ -164,7 +164,7 @@ static int sys_get_trip_temp(struct ther
 	if (thres_reg_value)
 		*temp = zonedev->tj_max - thres_reg_value * 1000;
 	else
-		*temp = 0;
+		*temp = THERMAL_TEMP_INVALID;
 	pr_debug("sys_get_trip_temp %d\n", *temp);
 
 	return 0;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #61 lore raw

PATCH 61/177 usb: dwc3: gadget: Properly track pending and queued SG

From: Thinh Nguyen <[email protected]>

commit 25dda9fc56bd90d45f9a4516bcfa5211e61b4290 upstream.

The driver incorrectly uses req->num_pending_sgs to track both the number of pending and queued SG entries. It only prepares the next request if the previous is done, and it doesn't update num_pending_sgs until there is TRB completion interrupt. This may starve the controller of more TRBs until the num_pending_sgs is decremented.

Fix this by decrementing the num_pending_sgs after they are queued and properly track both num_mapped_sgs and num_queued_sgs.

Fixes c96e6725db9d ("usb: dwc3: gadget: Correct the logic for queuing sgs")

Link https://lore.kernel.org/r/ba24591dbcaad8f244a3e88bd449bb7205a5aec3.1620874069.git.Thinh.Nguyen@synopsys.com open here

Reported-by Michael Grzeschik <[email protected]>

Tested-by Michael Grzeschik <[email protected]>

Acked-by Felipe Balbi <[email protected]>

Signed-off-by Thinh Nguyen <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/dwc3/gadget.c |   13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

drivers/usb/dwc3/gadget.c +7 -6

@@ -1162,6 +1162,7 @@ static void dwc3_prepare_one_trb_sg(stru
 			req->start_sg = sg_next(s);
 
 		req->num_queued_sgs++;
+		req->num_pending_sgs--;
 
 		/*
 		 * The number of pending SG entries may not correspond to the

@@ -1169,7 +1170,7 @@ static void dwc3_prepare_one_trb_sg(stru
 		 * don't include unused SG entries.
 		 */
 		if (length == 0) {
-			req->num_pending_sgs -= req->request.num_mapped_sgs - req->num_queued_sgs;
+			req->num_pending_sgs = 0;
 			break;
 		}

@@ -2602,15 +2603,15 @@ static int dwc3_gadget_ep_reclaim_trb_sg
 	struct dwc3_trb *trb = &dep->trb_pool[dep->trb_dequeue];
 	struct scatterlist *sg = req->sg;
 	struct scatterlist *s;
-	unsigned int pending = req->num_pending_sgs;
+	unsigned int num_queued = req->num_queued_sgs;
 	unsigned int i;
 	int ret = 0;
 
-	for_each_sg(sg, s, pending, i) {
+	for_each_sg(sg, s, num_queued, i) {
 		trb = &dep->trb_pool[dep->trb_dequeue];
 
 		req->sg = sg_next(s);
-		req->num_pending_sgs--;
+		req->num_queued_sgs--;
 
 		ret = dwc3_gadget_ep_reclaim_completed_trb(dep, req,
 				trb, event, status, true);

@@ -2633,7 +2634,7 @@ static int dwc3_gadget_ep_reclaim_trb_li
 
 static bool dwc3_gadget_ep_request_completed(struct dwc3_request *req)
 {
-	return req->num_pending_sgs == 0;
+	return req->num_pending_sgs == 0 && req->num_queued_sgs == 0;
 }
 
 static int dwc3_gadget_ep_cleanup_completed_request(struct dwc3_ep *dep,

@@ -2642,7 +2643,7 @@ static int dwc3_gadget_ep_cleanup_comple
 {
 	int ret;
 
-	if (req->num_pending_sgs)
+	if (req->request.num_mapped_sgs)
 		ret = dwc3_gadget_ep_reclaim_trb_sg(dep, req, event,
 				status);
 	else

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #62 lore raw

PATCH 62/177 usb: gadget: udc: renesas_usb3: Fix a race in usb3_start_pipen()

From: Yoshihiro Shimoda <[email protected]>

commit e752dbc59e1241b13b8c4f7b6eb582862e7668fe upstream.

The usb3_start_pipen() is called by renesas_usb3_ep_queue() and usb3_request_done_pipen() so that usb3_start_pipen() is possible to cause a race when getting usb3_first_req like below:

renesas_usb3_ep_queue()
spin_lock_irqsave()
list_add_tail()
spin_unlock_irqrestore()
usb3_start_pipen()
usb3_first_req = usb3_get_request() --- [1]
--- interrupt ---
usb3_irq_dma_int()
usb3_request_done_pipen()
usb3_get_request()
usb3_start_pipen()
usb3_first_req = usb3_get_request()
...
(the req is possible to be finished in the interrupt)

The usb3_first_req [1] above may have been finished after the interrupt ended so that this driver caused to start a transfer wrongly. To fix this issue, getting/checking the usb3_first_req are under spin_lock_irqsave() in the same section.

Fixes 746bfe63bba3 ("usb: gadget: renesas_usb3: add support for Renesas USB3.0 peripheral controller")

Cc stable <[email protected]>

Signed-off-by Yoshihiro Shimoda <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/gadget/udc/renesas_usb3.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

drivers/usb/gadget/udc/renesas_usb3.c +3 -2

@@ -1473,7 +1473,7 @@ static void usb3_start_pipen(struct rene
 			     struct renesas_usb3_request *usb3_req)
 {
 	struct renesas_usb3 *usb3 = usb3_ep_to_usb3(usb3_ep);
-	struct renesas_usb3_request *usb3_req_first = usb3_get_request(usb3_ep);
+	struct renesas_usb3_request *usb3_req_first;
 	unsigned long flags;
 	int ret = -EAGAIN;
 	u32 enable_bits = 0;

@@ -1481,7 +1481,8 @@ static void usb3_start_pipen(struct rene
 	spin_lock_irqsave(&usb3->lock, flags);
 	if (usb3_ep->halt || usb3_ep->started)
 		goto out;
-	if (usb3_req != usb3_req_first)
+	usb3_req_first = __usb3_get_request(usb3_ep);
+	if (!usb3_req_first || usb3_req != usb3_req_first)
 		goto out;
 
 	if (usb3_pn_change(usb3, usb3_ep->num) < 0)

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #63 lore raw

PATCH 63/177 net: usb: fix memory leak in smsc75xx_bind

From: Pavel Skripkin <[email protected]>

commit 46a8b29c6306d8bbfd92b614ef65a47c900d8e70 upstream.

Syzbot reported memory leak in smsc75xx_bind(). The problem was is non-freed memory in case of errors after memory allocation.

backtrace:

  [<ffffffff84245b62>] kmalloc include/linux/slab.h:556 [inline]
  [<ffffffff84245b62>] kzalloc include/linux/slab.h:686 [inline]
  [<ffffffff84245b62>] smsc75xx_bind+0x7a/0x334 drivers/net/usb/smsc75xx.c:1460
  [<ffffffff82b5b2e6>] usbnet_probe+0x3b6/0xc30 drivers/net/usb/usbnet.c:1728

Fixes d0cad871703b ("smsc75xx: SMSC LAN75xx USB gigabit ethernet adapter driver")

Reported-and-tested-by [email protected]

Signed-off-by Pavel Skripkin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/usb/smsc75xx.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

drivers/net/usb/smsc75xx.c +6 -2

@@ -1482,7 +1482,7 @@ static int smsc75xx_bind(struct usbnet *
 	ret = smsc75xx_wait_ready(dev, 0);
 	if (ret < 0) {
 		netdev_warn(dev->net, "device not ready in smsc75xx_bind\n");
-		return ret;
+		goto err;
 	}
 
 	smsc75xx_init_mac_address(dev);

@@ -1491,7 +1491,7 @@ static int smsc75xx_bind(struct usbnet *
 	ret = smsc75xx_reset(dev);
 	if (ret < 0) {
 		netdev_warn(dev->net, "smsc75xx_reset error %d\n", ret);
-		return ret;
+		goto err;
 	}
 
 	dev->net->netdev_ops = &smsc75xx_netdev_ops;

@@ -1501,6 +1501,10 @@ static int smsc75xx_bind(struct usbnet *
 	dev->hard_mtu = dev->net->mtu + dev->net->hard_header_len;
 	dev->net->max_mtu = MAX_SINGLE_PACKET_SIZE;
 	return 0;
+
+err:
+	kfree(pdata);
+	return ret;
 }
 
 static void smsc75xx_unbind(struct usbnet *dev, struct usb_interface *intf)

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #64 lore raw

PATCH 64/177 spi: spi-geni-qcom: Fix use-after-free on unbind

Link https://lore.kernel.org/r/dfa1d8c41b8acdfad87ec8654cd124e6e3cb3f31.1607286887.git.lukas@wunner.de open here

From: Lukas Wunner <[email protected]>

commit 8f96c434dfbc85ffa755d6634c8c1cb2233fcf24 upstream.

spi_geni_remove() accesses the driver's private data after calling spi_unregister_master() even though that function releases the last reference on the spi_master and thereby frees the private data.

Moreover, since commit 1a9e489e6128 ("spi: spi-geni-qcom: Use OPP API to set clk/perf state"), spi_geni_probe() leaks the spi_master allocation if the calls to dev_pm_opp_set_clkname() or dev_pm_opp_of_add_table() fail.

Fix by switching over to the new devm_spi_alloc_master() helper which keeps the private data accessible until the driver has unbound and also avoids the spi_master leak on probe.

Fixes 561de45f72bd ("spi: spi-geni-qcom: Add SPI driver support for GENI based QUP")

Signed-off-by Lukas Wunner <[email protected]>

Cc <[email protected]> # v4.20+: 5e844cc37a5c: spi: Introduce device-managed SPI controller allocation

Cc <[email protected]> # v4.20+

Cc Rajendra Nayak <[email protected]>

Cc Girish Mahadevan <[email protected]>

Signed-off-by Mark Brown <[email protected]>

[lukas: backport to v5.4.123]

Signed-off-by Lukas Wunner <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/spi/spi-geni-qcom.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

drivers/spi/spi-geni-qcom.c +1 -2

@@ -552,7 +552,7 @@ static int spi_geni_probe(struct platfor
 		return PTR_ERR(clk);
 	}
 
-	spi = spi_alloc_master(&pdev->dev, sizeof(*mas));
+	spi = devm_spi_alloc_master(&pdev->dev, sizeof(*mas));
 	if (!spi)
 		return -ENOMEM;

@@ -599,7 +599,6 @@ spi_geni_probe_free_irq:
 	free_irq(mas->irq, spi);
 spi_geni_probe_runtime_disable:
 	pm_runtime_disable(&pdev->dev);
-	spi_master_put(spi);
 	return ret;
 }

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #65 lore raw

PATCH 65/177 Bluetooth: cmtp: fix file refcount when cmtp_attach_device fails

From: Thadeu Lima de Souza Cascardo <[email protected]>

commit 8da3a0b87f4f1c3a3bbc4bfb78cf68476e97d183 upstream.

When cmtp_attach_device fails, cmtp_add_connection returns the error value which leads to the caller to doing fput through sockfd_put. But cmtp_session kthread, which is stopped in this path will also call fput, leading to a potential refcount underflow or a use-after-free.

Add a refcount before we signal the kthread to stop. The kthread will try to grab the cmtp_session_sem mutex before doing the fput, which is held when get_file is called, so there should be no races there.

Reported-by Ryota Shiga

Signed-off-by Thadeu Lima de Souza Cascardo <[email protected]>

Signed-off-by Marcel Holtmann <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/bluetooth/cmtp/core.c |    5 +++++
 1 file changed, 5 insertions(+)

net/bluetooth/cmtp/core.c +5 -0

@@ -392,6 +392,11 @@ int cmtp_add_connection(struct cmtp_conn
 	if (!(session->flags & BIT(CMTP_LOOPBACK))) {
 		err = cmtp_attach_device(session);
 		if (err < 0) {
+			/* Caller will call fput in case of failure, and so
+			 * will cmtp_session kthread.
+			 */
+			get_file(session->sock->file);
+
 			atomic_inc(&session->terminate);
 			wake_up_interruptible(sk_sleep(session->sock->sk));
 			up_write(&cmtp_session_sem);

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #66 lore raw

PATCH 66/177 fs/nfs: Use fatal_signal_pending instead of signal_pending

From: zhouchuangao <[email protected]>

commit bb002388901151fe35b6697ab116f6ed0721a9ed upstream.

We set the state of the current process to TASK_KILLABLE via prepare_to_wait(). Should we use fatal_signal_pending() to detect the signal here?

Fixes b4868b44c562 ("NFSv4: Wait for stateid updates after CLOSE/OPEN_DOWNGRADE")

Signed-off-by zhouchuangao <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/nfs/nfs4proc.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

fs/nfs/nfs4proc.c +2 -2

@@ -1647,7 +1647,7 @@ static void nfs_set_open_stateid_locked(
 		rcu_read_unlock();
 		trace_nfs4_open_stateid_update_wait(state->inode, stateid, 0);
 
-		if (!signal_pending(current)) {
+		if (!fatal_signal_pending(current)) {
 			if (schedule_timeout(5*HZ) == 0)
 				status = -EAGAIN;
 			else

@@ -3416,7 +3416,7 @@ static bool nfs4_refresh_open_old_statei
 		write_sequnlock(&state->seqlock);
 		trace_nfs4_close_stateid_update_wait(state->inode, dst, 0);
 
-		if (signal_pending(current))
+		if (fatal_signal_pending(current))
 			status = -EINTR;
 		else
 			if (schedule_timeout(5*HZ) != 0)

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #67 lore raw

PATCH 67/177 NFS: fix an incorrect limit in filelayout_decode_layout()

From: Dan Carpenter <[email protected]>

commit 769b01ea68b6c49dc3cde6adf7e53927dacbd3a8 upstream.

The "sizeof(struct nfs_fh)" is two bytes too large and could lead to memory corruption. It should be NFS_MAXFHSIZE because that's the size of the ->data[] buffer.

I reversed the size of the arguments to put the variable on the left.

Fixes 16b374ca439f ("NFSv4.1: pnfs: filelayout: add driver's LAYOUTGET and GETDEVICEINFO infrastructure")

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/nfs/filelayout/filelayout.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

fs/nfs/filelayout/filelayout.c +1 -1

@@ -717,7 +717,7 @@ filelayout_decode_layout(struct pnfs_lay
 		if (unlikely(!p))
 			goto out_err;
 		fl->fh_array[i]->size = be32_to_cpup(p++);
-		if (sizeof(struct nfs_fh) < fl->fh_array[i]->size) {
+		if (fl->fh_array[i]->size > NFS_MAXFHSIZE) {
 			printk(KERN_ERR "NFS: Too big fh %d received %d\n",
 			       i, fl->fh_array[i]->size);
 			goto out_err;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #68 lore raw

PATCH 68/177 NFS: Fix an Oopsable condition in __nfs_pageio_add_request()

From: Trond Myklebust <[email protected]>

commit 56517ab958b7c11030e626250c00b9b1a24b41eb upstream.

Ensure that nfs_pageio_error_cleanup() resets the mirror array contents, so that the structure reflects the fact that it is now empty. Also change the test in nfs_pageio_do_add_request() to be more robust by checking whether or not the list is empty rather than relying on the value of pg_count.

Fixes a7d42ddb3099 ("nfs: add mirroring support to pgio layer")

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/nfs/pagelist.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

fs/nfs/pagelist.c +5 -4

@@ -986,15 +986,16 @@ static int nfs_pageio_do_add_request(str
 
 	struct nfs_page *prev = NULL;
 
-	if (mirror->pg_count != 0) {
-		prev = nfs_list_entry(mirror->pg_list.prev);
-	} else {
+	if (list_empty(&mirror->pg_list)) {
 		if (desc->pg_ops->pg_init)
 			desc->pg_ops->pg_init(desc, req);
 		if (desc->pg_error < 0)
 			return 0;
 		mirror->pg_base = req->wb_pgbase;
-	}
+		mirror->pg_count = 0;
+		mirror->pg_recoalesce = 0;
+	} else
+		prev = nfs_list_entry(mirror->pg_list.prev);
 
 	if (desc->pg_maxretrans && req->wb_nio > desc->pg_maxretrans) {
 		if (NFS_SERVER(desc->pg_inode)->flags & NFS_MOUNT_SOFTERR)

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #69 lore raw

PATCH 69/177 NFS: Dont corrupt the value of pg_bytes_written in nfs_do_recoalesce()

From: Trond Myklebust <[email protected]>

commit 0d0ea309357dea0d85a82815f02157eb7fcda39f upstream.

The value of mirror->pg_bytes_written should only be updated after a successful attempt to flush out the requests on the list.

Fixes a7d42ddb3099 ("nfs: add mirroring support to pgio layer")

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/nfs/pagelist.c |   12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

fs/nfs/pagelist.c +5 -7

@@ -1019,17 +1019,16 @@ static void nfs_pageio_doio(struct nfs_p
 {
 	struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc);
 
-
 	if (!list_empty(&mirror->pg_list)) {
 		int error = desc->pg_ops->pg_doio(desc);
 		if (error < 0)
 			desc->pg_error = error;
-		else
+		if (list_empty(&mirror->pg_list)) {
 			mirror->pg_bytes_written += mirror->pg_count;
-	}
-	if (list_empty(&mirror->pg_list)) {
-		mirror->pg_count = 0;
-		mirror->pg_base = 0;
+			mirror->pg_count = 0;
+			mirror->pg_base = 0;
+			mirror->pg_recoalesce = 0;
+		}
 	}
 }

@@ -1123,7 +1122,6 @@ static int nfs_do_recoalesce(struct nfs_
 
 	do {
 		list_splice_init(&mirror->pg_list, &head);
-		mirror->pg_bytes_written -= mirror->pg_count;
 		mirror->pg_count = 0;
 		mirror->pg_base = 0;
 		mirror->pg_recoalesce = 0;

Greg Kroah-Hartman

author maintainer

2021-05-31 15:01 UTC #70 lore raw

PATCH 70/177 NFSv4: Fix v4.0/v4.1 SEEK_DATA return -ENOTSUPP when set NFS_V4_2 config

From: Zhang Xiaoxu <[email protected]>

commit e67afa7ee4a59584d7253e45d7f63b9528819a13 upstream.

Since commit bdcc2cd14e4e ("NFSv4.2: handle NFS-specific llseek errors"), nfs42_proc_llseek would return -EOPNOTSUPP rather than -ENOTSUPP when SEEK_DATA on NFSv4.0/v4.1.

This will lead xfstests generic/285 not run on NFSv4.0/v4.1 when set the CONFIG_NFS_V4_2, rather than run failed.

Fixes bdcc2cd14e4e ("NFSv4.2: handle NFS-specific llseek errors")

Cc <stable.vger.kernel.org> # 4.2

Signed-off-by Zhang Xiaoxu <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/nfs/nfs4file.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

fs/nfs/nfs4file.c +1 -1

@@ -168,7 +168,7 @@ static loff_t nfs4_file_llseek(struct fi
 	case SEEK_HOLE:
 	case SEEK_DATA:
 		ret = nfs42_proc_llseek(filep, offset, whence);
-		if (ret != -ENOTSUPP)
+		if (ret != -EOPNOTSUPP)
 			return ret;
 		/* Fall through */
 	default:

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #71 lore raw

PATCH 71/177 drm/meson: fix shutdown crash when component not probed

Link https://patchwork.freedesktop.org/patch/msgid/[email protected]

From: Neil Armstrong <[email protected]>

commit 7cfc4ea78fc103ea51ecbacd9236abb5b1c490d2 upstream.

When main component is not probed, by example when the dw-hdmi module is not loaded yet or in probe defer, the following crash appears on shutdown:

Unable to handle kernel NULL pointer dereference at virtual address 0000000000000038
...
pc : meson_drv_shutdown+0x24/0x50
lr : platform_drv_shutdown+0x20/0x30
...
Call trace:
meson_drv_shutdown+0x24/0x50
platform_drv_shutdown+0x20/0x30
device_shutdown+0x158/0x360
kernel_restart_prepare+0x38/0x48
kernel_restart+0x18/0x68
__do_sys_reboot+0x224/0x250
__arm64_sys_reboot+0x24/0x30

...

Simply check if the priv struct has been allocated before using it.

Fixes fa0c16caf3d7 ("drm: meson_drv add shutdown function")

Reported-by Stefan Agner <[email protected]>

Signed-off-by Neil Armstrong <[email protected]>

Tested-by Martin Blumenstingl <[email protected]>

Reviewed-by Martin Blumenstingl <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/gpu/drm/meson/meson_drv.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

drivers/gpu/drm/meson/meson_drv.c +5 -4

@@ -423,11 +423,12 @@ static int meson_probe_remote(struct pla
 static void meson_drv_shutdown(struct platform_device *pdev)
 {
 	struct meson_drm *priv = dev_get_drvdata(&pdev->dev);
-	struct drm_device *drm = priv->drm;
 
-	DRM_DEBUG_DRIVER("\n");
-	drm_kms_helper_poll_fini(drm);
-	drm_atomic_helper_shutdown(drm);
+	if (!priv)
+		return;
+
+	drm_kms_helper_poll_fini(priv->drm);
+	drm_atomic_helper_shutdown(priv->drm);
 }
 
 static int meson_drv_probe(struct platform_device *pdev)

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #72 lore raw

PATCH 72/177 net/mlx5e: Fix multipath lag activation

From: Dima Chumak <[email protected]>

commit 97817fcc684ed01497bd19d0cd4dea699665b9cf upstream.

When handling FIB_EVENT_ENTRY_REPLACE event for a new multipath route, lag activation can be missed if a stale (struct lag_mp)->mfi pointer exists, which was associated with an older multipath route that had been removed.

Normally, when a route is removed, it triggers mlx5_lag_fib_event(), which handles FIB_EVENT_ENTRY_DEL and clears mfi pointer. But, if mlx5_lag_check_prereq() condition isn't met, for example when eswitch is in legacy mode, the fib event is skipped and mfi pointer becomes stale.

Fix by resetting mfi pointer to NULL every time mlx5_lag_mp_init() is called.

Fixes 544fe7c2e654 ("net/mlx5e: Activate HW multipath and handle port affinity based on FIB events")

Signed-off-by Dima Chumak <[email protected]>

Reviewed-by Roi Dayan <[email protected]>

Signed-off-by Saeed Mahameed <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c |    6 ++++++
 1 file changed, 6 insertions(+)

drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c +6 -0

@@ -307,6 +307,11 @@ int mlx5_lag_mp_init(struct mlx5_lag *ld
 	struct lag_mp *mp = &ldev->lag_mp;
 	int err;
 
+	/* always clear mfi, as it might become stale when a route delete event
+	 * has been missed
+	 */
+	mp->mfi = NULL;
+
 	if (mp->fib_nb.notifier_call)
 		return 0;

@@ -328,4 +333,5 @@ void mlx5_lag_mp_cleanup(struct mlx5_lag
 
 	unregister_fib_notifier(&mp->fib_nb);
 	mp->fib_nb.notifier_call = NULL;
+	mp->mfi = NULL;
 }

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #73 lore raw

PATCH 73/177 net/mlx5e: Fix nullptr in add_vlan_push_action()

From: Dima Chumak <[email protected]>

commit dca59f4a791960ec73fa15803faa0abe0f92ece2 upstream.

The result of dev_get_by_index_rcu() is not checked for NULL and then gets dereferenced immediately.

Also, the RCU lock must be held by the caller of dev_get_by_index_rcu(), which isn't satisfied by the call stack.

Fix by handling nullptr return value when iflink device is not found. Add RCU locking around dev_get_by_index_rcu() to avoid possible adverse effects while iterating over the net_device's hlist.

It is safe not to increment reference count of the net_device pointer in case of a successful lookup, because it's already handled by VLAN code during VLAN device registration (see register_vlan_dev and netdev_upper_dev_link).

Fixes 278748a95aa3 ("net/mlx5e: Offload TC e-switch rules with egress VLAN device")

Addresses-Coverity: ("Dereference null return value")

Signed-off-by Dima Chumak <[email protected]>

Reviewed-by Vlad Buslov <[email protected]>

Signed-off-by Saeed Mahameed <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +6 -2

@@ -3170,8 +3170,12 @@ static int add_vlan_push_action(struct m
 	if (err)
 		return err;
 
-	*out_dev = dev_get_by_index_rcu(dev_net(vlan_dev),
-					dev_get_iflink(vlan_dev));
+	rcu_read_lock();
+	*out_dev = dev_get_by_index_rcu(dev_net(vlan_dev), dev_get_iflink(vlan_dev));
+	rcu_read_unlock();
+	if (!*out_dev)
+		return -ENODEV;
+
 	if (is_vlan_dev(*out_dev))
 		err = add_vlan_push_action(priv, attr, out_dev, action);

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #74 lore raw

PATCH 74/177 net/mlx4: Fix EEPROM dump support

From: Vladyslav Tarasiuk <[email protected]>

commit db825feefc6868896fed5e361787ba3bee2fd906 upstream.

Fix SFP and QSFP* EEPROM queries by setting i2c_address, offset and page number correctly. For SFP set the following params:
- I2C address for offsets 0-255 is 0x50. For 256-511 - 0x51.
- Page number is zero.
- Offset is 0-255.

At the same time, QSFP* parameters are different:
- I2C address is always 0x50.
- Page number is not limited to zero.
- Offset is 0-255 for page zero and 128-255 for others.

To set parameters accordingly to cable used, implement function to query module ID and implement respective helper functions to set parameters correctly.

Fixes 135dd9594f12 ("net/mlx4_en: ethtool, Remove unsupported SFP EEPROM high pages query")

Signed-off-by Vladyslav Tarasiuk <[email protected]>

Signed-off-by Tariq Toukan <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c |    4 
 drivers/net/ethernet/mellanox/mlx4/port.c       |  107 +++++++++++++++++++++++-
 2 files changed, 104 insertions(+), 7 deletions(-)

drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +1 -3

@@ -2011,8 +2011,6 @@ static int mlx4_en_set_tunable(struct ne
 	return ret;
 }
 
-#define MLX4_EEPROM_PAGE_LEN 256
-
 static int mlx4_en_get_module_info(struct net_device *dev,
 				   struct ethtool_modinfo *modinfo)
 {

@@ -2047,7 +2045,7 @@ static int mlx4_en_get_module_info(struc
 		break;
 	case MLX4_MODULE_ID_SFP:
 		modinfo->type = ETH_MODULE_SFF_8472;
-		modinfo->eeprom_len = MLX4_EEPROM_PAGE_LEN;
+		modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN;
 		break;
 	default:
 		return -EINVAL;

drivers/net/ethernet/mellanox/mlx4/port.c +103 -4

138 lines not on this page · raw message

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #75 lore raw

PATCH 75/177 Revert "net:tipc: Fix a double free in tipc_sk_mcast_rcv"

From: Hoang Le <[email protected]>

commit 75016891357a628d2b8acc09e2b9b2576c18d318 upstream.

This reverts commit 6bf24dc0cc0cc43b29ba344b66d78590e687e046. Above fix is not correct and caused memory leak issue.

Fixes 6bf24dc0cc0c ("net:tipc: Fix a double free in tipc_sk_mcast_rcv")

Acked-by Jon Maloy <[email protected]>

Acked-by Tung Nguyen <[email protected]>

Signed-off-by Hoang Le <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/tipc/socket.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

net/tipc/socket.c +4 -1

@@ -1210,7 +1210,10 @@ void tipc_sk_mcast_rcv(struct net *net,
 		spin_lock_bh(&inputq->lock);
 		if (skb_peek(arrvq) == skb) {
 			skb_queue_splice_tail_init(&tmpq, inputq);
-			__skb_dequeue(arrvq);
+			/* Decrease the skb's refcnt as increasing in the
+			 * function tipc_skb_peek
+			 */
+			kfree_skb(__skb_dequeue(arrvq));
 		}
 		spin_unlock_bh(&inputq->lock);
 		__skb_queue_purge(&tmpq);

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #76 lore raw

PATCH 76/177 tipc: wait and exit until all work queues are done

From: Xin Long <[email protected]>

commit 04c26faa51d1e2fe71cf13c45791f5174c37f986 upstream.

On some host, a crash could be triggered simply by repeating these commands several times:

  # modprobe tipc
  # tipc bearer enable media udp name UDP1 localip 127.0.0.1
  # rmmod tipc

[] BUG: unable to handle kernel paging request at ffffffffc096bb00 [] Workqueue: events 0xffffffffc096bb00
[] Call Trace:

  []  ? process_one_work+0x1a7/0x360
  []  ? worker_thread+0x30/0x390
  []  ? create_worker+0x1a0/0x1a0
  []  ? kthread+0x116/0x130
  []  ? kthread_flush_work_fn+0x10/0x10
  []  ? ret_from_fork+0x35/0x40

When removing the TIPC module, the UDP tunnel sock will be delayed to release in a work queue as sock_release() can't be done in rtnl_lock(). If the work queue is schedule to run after the TIPC module is removed, kernel will crash as the work queue function cleanup_beareri() code no longer exists when trying to invoke it.

To fix it, this patch introduce a member wq_count in tipc_net to track the numbers of work queues in schedule, and wait and exit until all work queues are done in tipc_exit_net().

Fixes d0f91938bede ("tipc: add ip/udp media type")

Reported-by Shuang Li <[email protected]>

Signed-off-by Xin Long <[email protected]>

Acked-by Jon Maloy <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/tipc/core.c      |    3 +++
 net/tipc/core.h      |    2 ++
 net/tipc/udp_media.c |    2 ++
 3 files changed, 7 insertions(+)

net/tipc/core.c +3 -0

@@ -107,6 +107,9 @@ static void __net_exit tipc_exit_net(str
 	tipc_bcast_stop(net);
 	tipc_nametbl_stop(net);
 	tipc_sk_rht_destroy(net);
+
+	while (atomic_read(&tn->wq_count))
+		cond_resched();
 }
 
 static struct pernet_operations tipc_net_ops = {

net/tipc/core.h +2 -0

@@ -143,6 +143,8 @@ struct tipc_net {
 
 	/* Work item for net finalize */
 	struct tipc_net_work final_work;
+	/* The numbers of work queues in schedule */
+	atomic_t wq_count;
 };
 
 static inline struct tipc_net *tipc_net(struct net *net)

net/tipc/udp_media.c +2 -0

@@ -802,6 +802,7 @@ static void cleanup_bearer(struct work_s
 		kfree_rcu(rcast, rcu);
 	}
 
+	atomic_dec(&tipc_net(sock_net(ub->ubsock->sk))->wq_count);
 	dst_cache_destroy(&ub->rcast.dst_cache);
 	udp_tunnel_sock_release(ub->ubsock);
 	synchronize_net();

@@ -822,6 +823,7 @@ static void tipc_udp_disable(struct tipc
 	RCU_INIT_POINTER(ub->bearer, NULL);
 
 	/* sock_release need to be done outside of rtnl lock */
+	atomic_inc(&tipc_net(sock_net(ub->ubsock->sk))->wq_count);
 	INIT_WORK(&ub->work, cleanup_bearer);
 	schedule_work(&ub->work);
 }

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #77 lore raw

PATCH 77/177 tipc: skb_linearize the head skb when reassembling msgs

From: Xin Long <[email protected]>

commit b7df21cf1b79ab7026f545e7bf837bd5750ac026 upstream.

It's not a good idea to append the frag skb to a skb's frag_list if the frag_list already has skbs from elsewhere, such as this skb was created by pskb_copy() where the frag_list was cloned (all the skbs in it were skb_get'ed) and shared by multiple skbs.

However, the new appended frag skb should have been only seen by the current skb. Otherwise, it will cause use after free crashes as this appended frag skb are seen by multiple skbs but it only got skb_get called once.

The same thing happens with a skb updated by pskb_may_pull() with a skb_cloned skb. Li Shuang has reported quite a few crashes caused by this when doing testing over macvlan devices:

[] kernel BUG at net/core/skbuff.c:1970!
[] Call Trace:

  []  skb_clone+0x4d/0xb0
  []  macvlan_broadcast+0xd8/0x160 [macvlan]
  []  macvlan_process_broadcast+0x148/0x150 [macvlan]
  []  process_one_work+0x1a7/0x360
  []  worker_thread+0x30/0x390

[] kernel BUG at mm/usercopy.c:102!
[] Call Trace:

  []  __check_heap_object+0xd3/0x100
  []  __check_object_size+0xff/0x16b
  []  simple_copy_to_iter+0x1c/0x30
  []  __skb_datagram_iter+0x7d/0x310
  []  __skb_datagram_iter+0x2a5/0x310
  []  skb_copy_datagram_iter+0x3b/0x90
  []  tipc_recvmsg+0x14a/0x3a0 [tipc]
  []  ____sys_recvmsg+0x91/0x150
  []  ___sys_recvmsg+0x7b/0xc0

[] kernel BUG at mm/slub.c:305!
[] Call Trace:

  []  <IRQ>
  []  kmem_cache_free+0x3ff/0x400
  []  __netif_receive_skb_core+0x12c/0xc40
  []  ? kmem_cache_alloc+0x12e/0x270
  []  netif_receive_skb_internal+0x3d/0xb0
  []  ? get_rx_page_info+0x8e/0xa0 [be2net]
  []  be_poll+0x6ef/0xd00 [be2net]
  []  ? irq_exit+0x4f/0x100
  []  net_rx_action+0x149/0x3b0

...

This patch is to fix it by linearizing the head skb if it has frag_list set in tipc_buf_append(). Note that we choose to do this before calling skb_unshare(), as __skb_linearize() will avoid skb_copy(). Also, we can not just drop the frag_list either as the early time.

Fixes 45c8b7b175ce ("tipc: allow non-linear first fragment buffer")

Reported-by Li Shuang <[email protected]>

Signed-off-by Xin Long <[email protected]>

Acked-by Jon Maloy <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/tipc/msg.c |    9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

net/tipc/msg.c +2 -7

@@ -141,18 +141,13 @@ int tipc_buf_append(struct sk_buff **hea
 		if (unlikely(head))
 			goto err;
 		*buf = NULL;
+		if (skb_has_frag_list(frag) && __skb_linearize(frag))
+			goto err;
 		frag = skb_unshare(frag, GFP_ATOMIC);
 		if (unlikely(!frag))
 			goto err;
 		head = *headbuf = frag;
 		TIPC_SKB_CB(head)->tail = NULL;
-		if (skb_is_nonlinear(head)) {
-			skb_walk_frags(head, tail) {
-				TIPC_SKB_CB(head)->tail = tail;
-			}
-		} else {
-			skb_frag_list_init(head);
-		}
 		return 0;
 	}

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #78 lore raw

PATCH 78/177 spi: spi-fsl-dspi: Fix a resource leak in an error handling path

Link https://lore.kernel.org/r/d51caaac747277a1099ba8dea07acd85435b857e.1620587472.git.christophe.jaillet@wanadoo.fr open here

From: Christophe JAILLET <[email protected]>

commit 680ec0549a055eb464dce6ffb4bfb736ef87236e upstream.

'dspi_request_dma()' should be undone by a 'dspi_release_dma()' call in the error handling path of the probe function, as already done in the remove function

Fixes 90ba37033cb9 ("spi: spi-fsl-dspi: Add DMA support for Vybrid")

Signed-off-by Christophe JAILLET <[email protected]>

Reviewed-by Vladimir Oltean <[email protected]>

Signed-off-by Mark Brown <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/spi/spi-fsl-dspi.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

drivers/spi/spi-fsl-dspi.c +3 -1

@@ -1142,11 +1142,13 @@ poll_mode:
 	ret = spi_register_controller(ctlr);
 	if (ret != 0) {
 		dev_err(&pdev->dev, "Problem registering DSPI ctlr\n");
-		goto out_free_irq;
+		goto out_release_dma;
 	}
 
 	return ret;
 
+out_release_dma:
+	dspi_release_dma(dspi);
 out_free_irq:
 	if (dspi->irq)
 		free_irq(dspi->irq, dspi);

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #79 lore raw

PATCH 79/177 net: dsa: mt7530: fix VLAN traffic leaks

From: DENG Qingfang <[email protected]>

commit 474a2ddaa192777522a7499784f1d60691cd831a upstream.

PCR_MATRIX field was set to all 1's when VLAN filtering is enabled, but was not reset when it is disabled, which may cause traffic leaks:

ip link add br0 type bridge vlan_filtering 1
ip link add br1 type bridge vlan_filtering 1
ip link set swp0 master br0
ip link set swp1 master br1
ip link set br0 type bridge vlan_filtering 0
ip link set br1 type bridge vlan_filtering 0
# traffic in br0 and br1 will start leaking to each other

As port_bridge_{add,del} have set up PCR_MATRIX properly, remove the PCR_MATRIX write from mt7530_port_set_vlan_aware.

Fixes 83163f7dca56 ("net: dsa: mediatek: add VLAN support for MT7530")

Signed-off-by DENG Qingfang <[email protected]>

Reviewed-by Florian Fainelli <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/dsa/mt7530.c |    8 --------
 1 file changed, 8 deletions(-)

drivers/net/dsa/mt7530.c +0 -8

@@ -809,14 +809,6 @@ mt7530_port_set_vlan_aware(struct dsa_sw
 {
 	struct mt7530_priv *priv = ds->priv;
 
-	/* The real fabric path would be decided on the membership in the
-	 * entry of VLAN table. PCR_MATRIX set up here with ALL_MEMBERS
-	 * means potential VLAN can be consisting of certain subset of all
-	 * ports.
-	 */
-	mt7530_rmw(priv, MT7530_PCR_P(port),
-		   PCR_MATRIX_MASK, PCR_MATRIX(MT7530_ALL_MEMBERS));
-
 	/* Trapped into security mode allows packet forwarding through VLAN
 	 * table lookup. CPU port is set to fallback mode to let untagged
 	 * frames pass through.

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #80 lore raw

PATCH 80/177 net: dsa: fix a crash if ->get_sset_count() fails

From: Dan Carpenter <[email protected]>

commit a269333fa5c0c8e53c92b5a28a6076a28cde3e83 upstream.

If ds->ops->get_sset_count() fails then it "count" is a negative error code such as -EOPNOTSUPP. Because "i" is an unsigned int, the negative error code is type promoted to a very high value and the loop will corrupt memory until the system crashes.

Fix this by checking for error codes and changing the type of "i" to just int.

Fixes badf3ada60ab ("net: dsa: Provide CPU port statistics to master netdev")

Reviewed-by Florian Fainelli <[email protected]>

Reviewed-by Vladimir Oltean <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/dsa/master.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

net/dsa/master.c +3 -2

@@ -147,8 +147,7 @@ static void dsa_master_get_strings(struc
 	struct dsa_switch *ds = cpu_dp->ds;
 	int port = cpu_dp->index;
 	int len = ETH_GSTRING_LEN;
-	int mcount = 0, count;
-	unsigned int i;
+	int mcount = 0, count, i;
 	uint8_t pfx[4];
 	uint8_t *ndata;

@@ -178,6 +177,8 @@ static void dsa_master_get_strings(struc
 		 */
 		ds->ops->get_strings(ds, port, stringset, ndata);
 		count = ds->ops->get_sset_count(ds, port, stringset);
+		if (count < 0)
+			return;
 		for (i = 0; i < count; i++) {
 			memmove(ndata + (i * len + sizeof(pfx)),
 				ndata + i * len, len - sizeof(pfx));

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #81 lore raw

PATCH 81/177 net: dsa: sja1105: error out on unsupported PHY mode

From: Vladimir Oltean <[email protected]>

commit 6729188d2646709941903052e4b78e1d82c239b9 upstream.

The driver continues probing when a port is configured for an unsupported PHY interface type, instead it should stop.

Fixes 8aa9ebccae87 ("net: dsa: Introduce driver for NXP SJA1105 5-port L2 switch")

Signed-off-by Vladimir Oltean <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/dsa/sja1105/sja1105_main.c |    1 +
 1 file changed, 1 insertion(+)

drivers/net/dsa/sja1105/sja1105_main.c +1 -0

@@ -178,6 +178,7 @@ static int sja1105_init_mii_settings(str
 		default:
 			dev_err(dev, "Unsupported PHY mode %s!\n",
 				phy_modes(ports[i].phy_mode));
+			return -EINVAL;
 		}
 
 		mii->phy_mac[i] = ports[i].role;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #82 lore raw

PATCH 82/177 i2c: s3c2410: fix possible NULL pointer deref on read message after write

From: Krzysztof Kozlowski <[email protected]>

commit 24990423267ec283b9d86f07f362b753eb9b0ed5 upstream.

Interrupt handler processes multiple message write requests one after another, till the driver message queue is drained. However if driver encounters a read message without preceding START, it stops the I2C transfer as it is an invalid condition for the controller. At least the comment describes a requirement "the controller forces us to send a new START when we change direction". This stop results in clearing the message queue (i2c->msg = NULL).

The code however immediately jumped back to label "retry_write" which dereferenced the "i2c->msg" making it a possible NULL pointer dereference.

The Coverity analysis:
1. Condition !is_msgend(i2c), taking false branch. if (!is_msgend(i2c)) {

2. Condition !is_lastmsg(i2c), taking true branch. } else if (!is_lastmsg(i2c)) {

3. Condition i2c->msg->flags & 1, taking true branch. if (i2c->msg->flags & I2C_M_RD) {

4. write_zero_model: Passing i2c to s3c24xx_i2c_stop, which sets i2c->msg to NULL. s3c24xx_i2c_stop(i2c, -EINVAL);

5. Jumping to label retry_write.
goto retry_write;

6. var_deref_model: Passing i2c to is_msgend, which dereferences null i2c->msg. if (!is_msgend(i2c)) {"

All previous calls to s3c24xx_i2c_stop() in this interrupt service routine are followed by jumping to end of function (acknowledging the interrupt and returning). This seems a reasonable choice also here since message buffer was entirely emptied.

Addresses-Coverity: Explicit null dereferenced

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")

Signed-off-by Krzysztof Kozlowski <[email protected]>

Signed-off-by Wolfram Sang <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/i2c/busses/i2c-s3c2410.c |    3 +++
 1 file changed, 3 insertions(+)

drivers/i2c/busses/i2c-s3c2410.c +3 -0

@@ -484,7 +484,10 @@ static int i2c_s3c_irq_nextbyte(struct s
 					 * forces us to send a new START
 					 * when we change direction
 					 */
+					dev_dbg(i2c->dev,
+						"missing START before write->read\n");
 					s3c24xx_i2c_stop(i2c, -EINVAL);
+					break;
 				}
 
 				goto retry_write;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #83 lore raw

PATCH 83/177 i2c: i801: Dont generate an interrupt on bus reset

From: Jean Delvare <[email protected]>

commit e4d8716c3dcec47f1557024add24e1f3c09eb24b upstream.

Now that the i2c-i801 driver supports interrupts, setting the KILL bit in a attempt to recover from a timed out transaction triggers an interrupt. Unfortunately, the interrupt handler (i801_isr) is not prepared for this situation and will try to process the interrupt as if it was signaling the end of a successful transaction. In the case of a block transaction, this can result in an out-of-range memory access.

This condition was reproduced several times by syzbot:
https://syzkaller.appspot.com/bug?extid=ed71512d469895b5b34e
https://syzkaller.appspot.com/bug?extid=8c8dedc0ba9e03f6c79e
https://syzkaller.appspot.com/bug?extid=c8ff0b6d6c73d81b610e
https://syzkaller.appspot.com/bug?extid=33f6c360821c399d69eb
https://syzkaller.appspot.com/bug?extid=be15dc0b1933f04b043a
https://syzkaller.appspot.com/bug?extid=b4d3fd1dfd53e90afd79

So disable interrupts while trying to reset the bus. Interrupts will be enabled again for the following transaction.

Fixes 636752bcb517 ("i2c-i801: Enable IRQ for SMBus transactions")

Reported-by [email protected]

Signed-off-by Jean Delvare <[email protected]>

Acked-by Andy Shevchenko <[email protected]>

Cc Jarkko Nikula <[email protected]>

Tested-by Jarkko Nikula <[email protected]>

Signed-off-by Wolfram Sang <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/i2c/busses/i2c-i801.c |    6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

drivers/i2c/busses/i2c-i801.c +2 -4

@@ -379,11 +379,9 @@ static int i801_check_post(struct i801_p
 		dev_err(&priv->pci_dev->dev, "Transaction timeout\n");
 		/* try to stop the current command */
 		dev_dbg(&priv->pci_dev->dev, "Terminating the current operation\n");
-		outb_p(inb_p(SMBHSTCNT(priv)) | SMBHSTCNT_KILL,
-		       SMBHSTCNT(priv));
+		outb_p(SMBHSTCNT_KILL, SMBHSTCNT(priv));
 		usleep_range(1000, 2000);
-		outb_p(inb_p(SMBHSTCNT(priv)) & (~SMBHSTCNT_KILL),
-		       SMBHSTCNT(priv));
+		outb_p(0, SMBHSTCNT(priv));
 
 		/* Check if it worked */
 		status = inb_p(SMBHSTSTS(priv));

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #84 lore raw

PATCH 84/177 i2c: sh_mobile: Use new clock calculation formulas for RZ/G2E

From: Geert Uytterhoeven <[email protected]>

commit c4740e293c93c747e65d53d9aacc2ba8521d1489 upstream.

When switching the Gen3 SoCs to the new clock calculation formulas, the match entry for RZ/G2E added in commit 51243b73455f2d12 ("i2c:
sh_mobile: Add support for r8a774c0 (RZ/G2E)") was forgotten.

Fixes e8a27567509b2439 ("i2c: sh_mobile: use new clock calculation formulas for Gen3")

Signed-off-by Geert Uytterhoeven <[email protected]>

Reviewed-by Fabrizio Castro <[email protected]>

Signed-off-by Wolfram Sang <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/i2c/busses/i2c-sh_mobile.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/i2c/busses/i2c-sh_mobile.c +1 -1

@@ -813,7 +813,7 @@ static const struct sh_mobile_dt_config
 static const struct of_device_id sh_mobile_i2c_dt_ids[] = {
 	{ .compatible = "renesas,iic-r8a73a4", .data = &fast_clock_dt_config },
 	{ .compatible = "renesas,iic-r8a7740", .data = &r8a7740_dt_config },
-	{ .compatible = "renesas,iic-r8a774c0", .data = &fast_clock_dt_config },
+	{ .compatible = "renesas,iic-r8a774c0", .data = &v2_freq_calc_dt_config },
 	{ .compatible = "renesas,iic-r8a7790", .data = &v2_freq_calc_dt_config },
 	{ .compatible = "renesas,iic-r8a7791", .data = &v2_freq_calc_dt_config },
 	{ .compatible = "renesas,iic-r8a7792", .data = &v2_freq_calc_dt_config },

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #85 lore raw

PATCH 85/177 perf jevents: Fix getting maximum number of fds

Link http://lore.kernel.org/lkml/[email protected] open here

From: Felix Fietkau <[email protected]>

commit 75ea44e356b5de8c817f821c9dd68ae329e82add upstream.

On some hosts, rlim.rlim_max can be returned as RLIM_INFINITY. By casting it to int, it is interpreted as -1, which will cause get_maxfds to return 0, causing "Invalid argument" errors in nftw() calls. Fix this by casting the second argument of min() to rlim_t instead.

Fixes 80eeb67fe577 ("perf jevents: Program to convert JSON file")

Signed-off-by Felix Fietkau <[email protected]>

Cc Alexander Shishkin <[email protected]>

Cc Andi Kleen <[email protected]>

Cc Jiri Olsa <[email protected]>

Cc Mark Rutland <[email protected]>

Cc Namhyung Kim <[email protected]>

Cc Peter Zijlstra <[email protected]>

Cc Sukadev Bhattiprolu <[email protected]>

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/perf/pmu-events/jevents.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

tools/perf/pmu-events/jevents.c +1 -1

@@ -862,7 +862,7 @@ static int get_maxfds(void)
 	struct rlimit rlim;
 
 	if (getrlimit(RLIMIT_NOFILE, &rlim) == 0)
-		return min((int)rlim.rlim_max / 2, 512);
+		return min(rlim.rlim_max / 2, (rlim_t)512);
 
 	return 512;
 }

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #86 lore raw

PATCH 86/177 platform/x86: hp_accel: Avoid invoking _INI to speed up resume

From: Kai-Heng Feng <[email protected]>

[ Upstream commit 79d341e26ebcdbc622348aaaab6f8f89b6fdb25f ]

hp_accel can take almost two seconds to resume on some HP laptops.

The bottleneck is on evaluating _INI, which is only needed to run once.

Resolve the issue by only invoking _INI when it's necessary. Namely, on probe and on hibernation restore.

Signed-off-by Kai-Heng Feng <[email protected]>

Acked-by Éric Piel <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/misc/lis3lv02d/lis3lv02d.h |  1 +
 drivers/platform/x86/hp_accel.c    | 22 +++++++++++++++++++++-
 2 files changed, 22 insertions(+), 1 deletion(-)

drivers/misc/lis3lv02d/lis3lv02d.h +1 -0

index 1b0c99883c57..c008eecfdfe8 100644
@@ -271,6 +271,7 @@ struct lis3lv02d {
 	int			regs_size;
 	u8                      *reg_cache;
 	bool			regs_stored;
+	bool			init_required;
 	u8                      odr_mask;  /* ODR bit mask */
 	u8			whoami;    /* indicates measurement precision */
 	s16 (*read_data) (struct lis3lv02d *lis3, int reg);

drivers/platform/x86/hp_accel.c +21 -1

index 799cbe2ffcf3..8c0867bda828 100644
@@ -88,6 +88,9 @@ MODULE_DEVICE_TABLE(acpi, lis3lv02d_device_ids);
 static int lis3lv02d_acpi_init(struct lis3lv02d *lis3)
 {
 	struct acpi_device *dev = lis3->bus_priv;
+	if (!lis3->init_required)
+		return 0;
+
 	if (acpi_evaluate_object(dev->handle, METHOD_NAME__INI,
 				 NULL, NULL) != AE_OK)
 		return -EINVAL;

@@ -356,6 +359,7 @@ static int lis3lv02d_add(struct acpi_device *device)
 	}
 
 	/* call the core layer do its init */
+	lis3_dev.init_required = true;
 	ret = lis3lv02d_init_device(&lis3_dev);
 	if (ret)
 		return ret;

@@ -403,11 +407,27 @@ static int lis3lv02d_suspend(struct device *dev)
 
 static int lis3lv02d_resume(struct device *dev)
 {
+	lis3_dev.init_required = false;
+	lis3lv02d_poweron(&lis3_dev);
+	return 0;
+}
+
+static int lis3lv02d_restore(struct device *dev)
+{
+	lis3_dev.init_required = true;
 	lis3lv02d_poweron(&lis3_dev);
 	return 0;
 }
 
-static SIMPLE_DEV_PM_OPS(hp_accel_pm, lis3lv02d_suspend, lis3lv02d_resume);
+static const struct dev_pm_ops hp_accel_pm = {
+	.suspend = lis3lv02d_suspend,
+	.resume = lis3lv02d_resume,
+	.freeze = lis3lv02d_suspend,
+	.thaw = lis3lv02d_resume,
+	.poweroff = lis3lv02d_suspend,
+	.restore = lis3lv02d_restore,
+};
+
 #define HP_ACCEL_PM (&hp_accel_pm)
 #else
 #define HP_ACCEL_PM NULL

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #87 lore raw

PATCH 87/177 gpio: cadence: Add missing MODULE_DEVICE_TABLE

From: Zou Wei <[email protected]>

[ Upstream commit 1e948b1752b58c9c570989ab29ceef5b38fdccda ]

This patch adds missing MODULE_DEVICE_TABLE definition which generates correct modalias for automatic loading of this driver when it is built as an external module.

Reported-by Hulk Robot <[email protected]>

Signed-off-by Zou Wei <[email protected]>

Signed-off-by Bartosz Golaszewski <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpio/gpio-cadence.c | 1 +
 1 file changed, 1 insertion(+)

drivers/gpio/gpio-cadence.c +1 -0

index a4d3239d2594..4ab3fcd9b9ba 100644
@@ -278,6 +278,7 @@ static const struct of_device_id cdns_of_ids[] = {
 	{ .compatible = "cdns,gpio-r1p02" },
 	{ /* sentinel */ },
 };
+MODULE_DEVICE_TABLE(of, cdns_of_ids);
 
 static struct platform_driver cdns_gpio_driver = {
 	.driver = {

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #88 lore raw

PATCH 88/177 Revert "media: usb: gspca: add a missed check for goto_low_power"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit fd013265e5b5576a74a033920d6c571e08d7c423 ]

This reverts commit 5b711870bec4dc9a6d705d41e127e73944fa3650.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to do does nothing useful as a user can do nothing with this information and if an error did happen, the code would continue on as before. Because of this, just revert it.

Cc Kangjie Lu <[email protected]>

Cc Mauro Carvalho Chehab <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/media/usb/gspca/cpia1.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

drivers/media/usb/gspca/cpia1.c +1 -5

index a4f7431486f3..d93d384286c1 100644
@@ -1424,7 +1424,6 @@ static int sd_config(struct gspca_dev *gspca_dev,
 {
 	struct sd *sd = (struct sd *) gspca_dev;
 	struct cam *cam;
-	int ret;
 
 	sd->mainsFreq = FREQ_DEF == V4L2_CID_POWER_LINE_FREQUENCY_60HZ;
 	reset_camera_params(gspca_dev);

@@ -1436,10 +1435,7 @@ static int sd_config(struct gspca_dev *gspca_dev,
 	cam->cam_mode = mode;
 	cam->nmodes = ARRAY_SIZE(mode);
 
-	ret = goto_low_power(gspca_dev);
-	if (ret)
-		gspca_err(gspca_dev, "Cannot go to low power mode: %d\n",
-			  ret);
+	goto_low_power(gspca_dev);
 	/* Check the firmware version. */
 	sd->params.version.firmwareVersion = 0;
 	get_version_information(gspca_dev);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #89 lore raw

PATCH 89/177 Revert "ALSA: sb: fix a missing check of snd_ctl_add"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 4b059ce1f4b368208c2310925f49be77f15e527b ]

This reverts commit beae77170c60aa786f3e4599c18ead2854d8694d.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It is safe to ignore this error as the mixer element is optional, and the driver is very legacy.

Cc Aditya Pakki <[email protected]>

Reviewed-by Takashi Iwai <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 sound/isa/sb/sb16_main.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

sound/isa/sb/sb16_main.c +3 -7

index 0768bbf8fd71..679f9f48370f 100644
@@ -864,14 +864,10 @@ int snd_sb16dsp_pcm(struct snd_sb *chip, int device)
 	snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK, &snd_sb16_playback_ops);
 	snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_CAPTURE, &snd_sb16_capture_ops);
 
-	if (chip->dma16 >= 0 && chip->dma8 != chip->dma16) {
-		err = snd_ctl_add(card, snd_ctl_new1(
-					&snd_sb16_dma_control, chip));
-		if (err)
-			return err;
-	} else {
+	if (chip->dma16 >= 0 && chip->dma8 != chip->dma16)
+		snd_ctl_add(card, snd_ctl_new1(&snd_sb16_dma_control, chip));
+	else
 		pcm->info_flags = SNDRV_PCM_INFO_HALF_DUPLEX;
-	}
 
 	snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV,
 					      card->dev,

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #90 lore raw

PATCH 90/177 Revert "serial: max310x: pass return value of spi_register_driver"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit b0a85abbe92e1a6f3e8580a4590fa7245de7090b ]

This reverts commit 51f689cc11333944c7a457f25ec75fcb41e99410.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

This change did not properly unwind from the error condition, so it was not correct.

Cc Kangjie Lu <[email protected]>

Acked-by Jiri Slaby <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/tty/serial/max310x.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/tty/serial/max310x.c +2 -2

index 8434bd5a8ec7..f60b7b86d099 100644
@@ -1527,10 +1527,10 @@ static int __init max310x_uart_init(void)
 		return ret;
 
 #ifdef CONFIG_SPI_MASTER
-	ret = spi_register_driver(&max310x_spi_driver);
+	spi_register_driver(&max310x_spi_driver);
 #endif
 
-	return ret;
+	return 0;
 }
 module_init(max310x_uart_init);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #91 lore raw

PATCH 91/177 serial: max310x: unregister uart driver in case of failure and abort

From: Atul Gopinathan <[email protected]>

[ Upstream commit 3890e3dea315f1a257d1b940a2a4e2fa16a7b095 ]

The macro "spi_register_driver" invokes the function "__spi_register_driver()" which has a return type of int and can fail, returning a negative value in such a case. This is currently ignored and the init() function yields success even if the spi driver failed to register.

Fix this by collecting the return value of "__spi_register_driver()" and also unregister the uart driver in case of failure.

Cc Jiri Slaby <[email protected]>

Signed-off-by Atul Gopinathan <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/tty/serial/max310x.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

drivers/tty/serial/max310x.c +4 -2

index f60b7b86d099..5bf8dd6198bb 100644
@@ -1527,10 +1527,12 @@ static int __init max310x_uart_init(void)
 		return ret;
 
 #ifdef CONFIG_SPI_MASTER
-	spi_register_driver(&max310x_spi_driver);
+	ret = spi_register_driver(&max310x_spi_driver);
+	if (ret)
+		uart_unregister_driver(&max310x_uart);
 #endif
 
-	return 0;
+	return ret;
 }
 module_init(max310x_uart_init);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #92 lore raw

PATCH 92/177 Revert "net: fujitsu: fix a potential NULL pointer dereference"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 5f94eaa4ee23e80841fa359a372f84cfe25daee1 ]

This reverts commit 9f4d6358e11bbc7b839f9419636188e4151fb6e4.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The original change does not change any behavior as the caller of this function onlyu checks for "== -1" as an error condition so this error is not handled properly. Remove this change and it will be fixed up properly in a later commit.

Cc Kangjie Lu <[email protected]>

Cc David S. Miller <[email protected]>

Reviewed-by Dominik Brodowski <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/fujitsu/fmvj18x_cs.c | 5 -----
 1 file changed, 5 deletions(-)

drivers/net/ethernet/fujitsu/fmvj18x_cs.c +0 -5

index 1eca0fdb9933..a69cd19a55ae 100644
@@ -547,11 +547,6 @@ static int fmvj18x_get_hwinfo(struct pcmcia_device *link, u_char *node_id)
 	return -1;
 
     base = ioremap(link->resource[2]->start, resource_size(link->resource[2]));
-    if (!base) {
-	    pcmcia_release_window(link, link->resource[2]);
-	    return -ENOMEM;
-    }
-
     pcmcia_map_mem_page(link, link->resource[2], 0);
 
     /*

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #93 lore raw

PATCH 93/177 net: fujitsu: fix potential null-ptr-deref

From: Anirudh Rayabharam <[email protected]>

[ Upstream commit 52202be1cd996cde6e8969a128dc27ee45a7cb5e ]

In fmvj18x_get_hwinfo(), if ioremap fails there will be NULL pointer deref. To fix this, check the return value of ioremap and return -1 to the caller in case of failure.

Cc "David S. Miller" <[email protected]>

Acked-by Dominik Brodowski <[email protected]>

Signed-off-by Anirudh Rayabharam <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/fujitsu/fmvj18x_cs.c | 5 +++++
 1 file changed, 5 insertions(+)

drivers/net/ethernet/fujitsu/fmvj18x_cs.c +5 -0

index a69cd19a55ae..b8fc9bbeca2c 100644
@@ -547,6 +547,11 @@ static int fmvj18x_get_hwinfo(struct pcmcia_device *link, u_char *node_id)
 	return -1;
 
     base = ioremap(link->resource[2]->start, resource_size(link->resource[2]));
+    if (!base) {
+	pcmcia_release_window(link, link->resource[2]);
+	return -1;
+    }
+
     pcmcia_map_mem_page(link, link->resource[2], 0);
 
     /*

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #94 lore raw

PATCH 94/177 Revert "net/smc: fix a NULL pointer dereference"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 5369ead83f5aff223b6418c99cb1fe9a8f007363 ]

This reverts commit e183d4e414b64711baf7a04e214b61969ca08dfa.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The original commit causes a memory leak and does not properly fix the issue it claims to fix. I will send a follow-on patch to resolve this properly.

Cc Kangjie Lu <[email protected]>

Cc Ursula Braun <[email protected]>

Cc David S. Miller <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/smc/smc_ism.c | 5 -----
 1 file changed, 5 deletions(-)

net/smc/smc_ism.c +0 -5

index e89e918b88e0..2fff79db1a59 100644
@@ -289,11 +289,6 @@ struct smcd_dev *smcd_alloc_dev(struct device *parent, const char *name,
 	INIT_LIST_HEAD(&smcd->vlan);
 	smcd->event_wq = alloc_ordered_workqueue("ism_evt_wq-%s)",
 						 WQ_MEM_RECLAIM, name);
-	if (!smcd->event_wq) {
-		kfree(smcd->conn);
-		kfree(smcd);
-		return NULL;
-	}
 	return smcd;
 }
 EXPORT_SYMBOL_GPL(smcd_alloc_dev);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #95 lore raw

PATCH 95/177 net: caif: remove BUG_ON(dev == NULL) in caif_xmit

From: Du Cheng <[email protected]>

[ Upstream commit 65a67792e3416f7c5d7daa47d99334cbb19a7449 ]

The condition of dev == NULL is impossible in caif_xmit(), hence it is for the removal.

Explanation:
The static caif_xmit() is only called upon via a function pointer ndo_start_xmit defined in include/linux/netdevice.h:

```
struct net_device_ops {
    ...
    netdev_tx_t     (*ndo_start_xmit)(struct sk_buff *skb, struct net_device *dev);
    ...
}
```

The exhausive list of call points are:

```
drivers/net/ethernet/qualcomm/rmnet/rmnet_map_command.c
    dev->netdev_ops->ndo_start_xmit(skb, dev);
    ^                                    ^

drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c
    struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev);
			     ^                       ^
    return adapter->rn_ops->ndo_start_xmit(skb, netdev); // adapter would crash first
	   ^                                    ^

drivers/usb/gadget/function/f_ncm.c
    ncm->netdev->netdev_ops->ndo_start_xmit(NULL, ncm->netdev);
	      ^                                   ^

include/linux/netdevice.h
static inline netdev_tx_t __netdev_start_xmit(...
{
    return ops->ndo_start_xmit(skb, dev);
				    ^
}

    const struct net_device_ops *ops = dev->netdev_ops;
				       ^
    rc = __netdev_start_xmit(ops, skb, dev, more);
				       ^
```

In each of the enumerated scenarios, it is impossible for the NULL-valued dev to reach the caif_xmit() without crashing the kernel earlier, therefore BUG_ON(dev == NULL) is rather useless, hence the removal.

Cc David S. Miller <[email protected]>

Signed-off-by Du Cheng <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/caif/caif_serial.c | 1 -
 1 file changed, 1 deletion(-)

drivers/net/caif/caif_serial.c +0 -1

index 40b079162804..0f2bee59a82b 100644
@@ -270,7 +270,6 @@ static int caif_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ser_device *ser;
 
-	BUG_ON(dev == NULL);
 	ser = netdev_priv(dev);
 
 	/* Send flow off once, on high water mark */

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #96 lore raw

PATCH 96/177 Revert "char: hpet: fix a missing check of ioremap"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 566f53238da74801b48e985788e5f7c9159e5940 ]

This reverts commit 13bd14a41ce3105d5b1f3cd8b4d1e249d17b6d9b.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

While this is technically correct, it is only fixing ONE of these errors in this function, so the patch is not fully correct. I'll leave this revert and provide a fix for this later that resolves this same "problem" everywhere in this function.

Cc Kangjie Lu <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/char/hpet.c | 2 --
 1 file changed, 2 deletions(-)

drivers/char/hpet.c +0 -2

index f69609b47fef..3e31740444f1 100644
@@ -969,8 +969,6 @@ static acpi_status hpet_resources(struct acpi_resource *res, void *data)
 	if (ACPI_SUCCESS(status)) {
 		hdp->hd_phys_address = addr.address.minimum;
 		hdp->hd_address = ioremap(addr.address.minimum, addr.address.address_length);
-		if (!hdp->hd_address)
-			return AE_ERROR;
 
 		if (hpet_is_known(hdp)) {
 			iounmap(hdp->hd_address);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:21 UTC #97 lore raw

PATCH 97/177 char: hpet: add checks after calling ioremap

From: Tom Seewald <[email protected]>

[ Upstream commit b11701c933112d49b808dee01cb7ff854ba6a77a ]

The function hpet_resources() calls ioremap() two times, but in both cases it does not check if ioremap() returned a null pointer. Fix this by adding null pointer checks and returning an appropriate error.

Signed-off-by Tom Seewald <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/char/hpet.c | 4 ++++
 1 file changed, 4 insertions(+)

drivers/char/hpet.c +4 -0

index 3e31740444f1..d390ab5e51d3 100644
@@ -969,6 +969,8 @@ static acpi_status hpet_resources(struct acpi_resource *res, void *data)
 	if (ACPI_SUCCESS(status)) {
 		hdp->hd_phys_address = addr.address.minimum;
 		hdp->hd_address = ioremap(addr.address.minimum, addr.address.address_length);
+		if (!hdp->hd_address)
+			return AE_ERROR;
 
 		if (hpet_is_known(hdp)) {
 			iounmap(hdp->hd_address);

@@ -982,6 +984,8 @@ static acpi_status hpet_resources(struct acpi_resource *res, void *data)
 		hdp->hd_phys_address = fixmem32->address;
 		hdp->hd_address = ioremap(fixmem32->address,
 						HPET_RANGE_SIZE);
+		if (!hdp->hd_address)
+			return AE_ERROR;
 
 		if (hpet_is_known(hdp)) {
 			iounmap(hdp->hd_address);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #98 lore raw

PATCH 98/177 Revert "ALSA: gus: add a check of the status of snd_ctl_add"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 1dacca7fa1ebea47d38d20cd2df37094805d2649 ]

This reverts commit 0f25e000cb4398081748e54f62a902098aa79ec1.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The original commit did nothing if there was an error, except to print out a message, which is pointless. So remove the commit as it gives a "false sense of doing something".

Cc Kangjie Lu <[email protected]>

Reviewed-by Takashi Iwai <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 sound/isa/gus/gus_main.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

sound/isa/gus/gus_main.c +2 ~~-11~~

index af6b4d89d695..39911a637e80 100644
@@ -77,17 +77,8 @@ static const struct snd_kcontrol_new snd_gus_joystick_control = {
 
 static void snd_gus_init_control(struct snd_gus_card *gus)
 {
-	int ret;
-
-	if (!gus->ace_flag) {
-		ret =
-			snd_ctl_add(gus->card,
-					snd_ctl_new1(&snd_gus_joystick_control,
-						gus));
-		if (ret)
-			snd_printk(KERN_ERR "gus: snd_ctl_add failed: %d\n",
-					ret);
-	}
+	if (!gus->ace_flag)
+		snd_ctl_add(gus->card, snd_ctl_new1(&snd_gus_joystick_control, gus));
 }
 
 /*

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #99 lore raw

PATCH 99/177 ALSA: sb8: Add a comment note regarding an unused pointer

From: Atul Gopinathan <[email protected]>

[ Upstream commit a28591f61b60fac820c6de59826ffa710e5e314e ]

The field "fm_res" of "struct snd_sb8" is never used/dereferenced throughout the sb8.c code. Therefore there is no need for any null value check after the "request_region()".

Add a comment note to make developers know about this and prevent any "NULL check" patches on this part of code.

Cc Takashi Iwai <[email protected]>

Signed-off-by Atul Gopinathan <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 sound/isa/sb/sb8.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

sound/isa/sb/sb8.c +5 -1

index 6f1fc8789676..4386c24d2dcc 100644
@@ -94,7 +94,11 @@ static int snd_sb8_probe(struct device *pdev, unsigned int dev)
 	acard = card->private_data;
 	card->private_free = snd_sb8_free;
 
-	/* block the 0x388 port to avoid PnP conflicts */
+	/*
+	 * Block the 0x388 port to avoid PnP conflicts.
+	 * No need to check this value after request_region,
+	 * as we never do anything with it.
+	 */
 	acard->fm_res = request_region(0x388, 4, "SoundBlaster FM");
 
 	if (port[dev] != SNDRV_AUTO_PORT) {

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:22 UTC #100 lore raw

PATCH 100/177 Revert "ALSA: usx2y: Fix potential NULL pointer dereference"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 4667a6fc1777ce071504bab570d3599107f4790f ]

This reverts commit a2c6433ee5a35a8de6d563f6512a26f87835ea0f.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The original patch was incorrect, and would leak memory if the error path the patch added was hit.

Cc Aditya Pakki <[email protected]>

Reviewed-by Takashi Iwai <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 sound/usb/usx2y/usb_stream.c | 5 -----
 1 file changed, 5 deletions(-)

sound/usb/usx2y/usb_stream.c +0 -5

index 091c071b270a..6bba17bf689a 100644
@@ -91,12 +91,7 @@ static int init_urbs(struct usb_stream_kernel *sk, unsigned use_packsize,
 
 	for (u = 0; u < USB_STREAM_NURBS; ++u) {
 		sk->inurb[u] = usb_alloc_urb(sk->n_o_ps, GFP_KERNEL);
-		if (!sk->inurb[u])
-			return -ENOMEM;
-
 		sk->outurb[u] = usb_alloc_urb(sk->n_o_ps, GFP_KERNEL);
-		if (!sk->outurb[u])
-			return -ENOMEM;
 	}
 
 	if (init_pipe_urbs(sk, use_packsize, sk->inurb, indata, dev, in_pipe) ||

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #101 lore raw

PATCH 101/177 Revert "isdn: mISDNinfineon: fix potential NULL pointer dereference"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit abd7bca23bd4247124265152d00ffd4b2b0d6877 ]

This reverts commit d721fe99f6ada070ae8fc0ec3e01ce5a42def0d9.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The original commit was incorrect, it should have never have used "unlikely()" and if it ever does trigger, resources are left grabbed.

Given there are no users for this code around, I'll just revert this and leave it "as is" as the odds that ioremap() will ever fail here is horrendiously low.

Cc Kangjie Lu <[email protected]>

Cc David S. Miller <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/isdn/hardware/mISDN/mISDNinfineon.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

drivers/isdn/hardware/mISDN/mISDNinfineon.c +1 -4

index f4cb29766888..d62006bab9c6 100644
@@ -697,11 +697,8 @@ setup_io(struct inf_hw *hw)
 				(ulong)hw->addr.start, (ulong)hw->addr.size);
 			return err;
 		}
-		if (hw->ci->addr_mode == AM_MEMIO) {
+		if (hw->ci->addr_mode == AM_MEMIO)
 			hw->addr.p = ioremap(hw->addr.start, hw->addr.size);
-			if (unlikely(!hw->addr.p))
-				return -ENOMEM;
-		}
 		hw->addr.mode = hw->ci->addr_mode;
 		if (debug & DEBUG_HW)
 			pr_notice("%s: IO addr %lx (%lu bytes) mode%d\n",

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #102 lore raw

PATCH 102/177 isdn: mISDNinfineon: check/cleanup ioremap failure correctly in setup_io

From: Phillip Potter <[email protected]>

[ Upstream commit c446f0d4702d316e1c6bf621f70e79678d28830a ]

Move hw->cfg.mode and hw->addr.mode assignments from hw->ci->cfg_mode and hw->ci->addr_mode respectively, to be before the subsequent checks for memory IO mode (and possible ioremap calls in this case).

Also introduce ioremap error checks at both locations. This allows resources to be properly freed on ioremap failure, as when the caller of setup_io then subsequently calls release_io via its error path, release_io can now correctly determine the mode as it has been set before the ioremap call.

Finally, refactor release_io function so that it will call release_mem_region in the memory IO case, regardless of whether or not hw->cfg.p/hw->addr.p are NULL. This means resources are then properly released on failure.

This properly implements the original reverted commit (d721fe99f6ad) from the University of Minnesota, whilst also implementing the ioremap check for the hw->ci->cfg_mode if block as well.

Cc David S. Miller <[email protected]>

Signed-off-by Phillip Potter <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/isdn/hardware/mISDN/mISDNinfineon.c | 24 ++++++++++++++-------
 1 file changed, 16 insertions(+), 8 deletions(-)

drivers/isdn/hardware/mISDN/mISDNinfineon.c +16 -8

index d62006bab9c6..3cf0c6f5a1dc 100644
@@ -630,17 +630,19 @@ static void
 release_io(struct inf_hw *hw)
 {
 	if (hw->cfg.mode) {
-		if (hw->cfg.p) {
+		if (hw->cfg.mode == AM_MEMIO) {
 			release_mem_region(hw->cfg.start, hw->cfg.size);
-			iounmap(hw->cfg.p);
+			if (hw->cfg.p)
+				iounmap(hw->cfg.p);
 		} else
 			release_region(hw->cfg.start, hw->cfg.size);
 		hw->cfg.mode = AM_NONE;
 	}
 	if (hw->addr.mode) {
-		if (hw->addr.p) {
+		if (hw->addr.mode == AM_MEMIO) {
 			release_mem_region(hw->addr.start, hw->addr.size);
-			iounmap(hw->addr.p);
+			if (hw->addr.p)
+				iounmap(hw->addr.p);
 		} else
 			release_region(hw->addr.start, hw->addr.size);
 		hw->addr.mode = AM_NONE;

@@ -670,9 +672,12 @@ setup_io(struct inf_hw *hw)
 				(ulong)hw->cfg.start, (ulong)hw->cfg.size);
 			return err;
 		}
-		if (hw->ci->cfg_mode == AM_MEMIO)
-			hw->cfg.p = ioremap(hw->cfg.start, hw->cfg.size);
 		hw->cfg.mode = hw->ci->cfg_mode;
+		if (hw->ci->cfg_mode == AM_MEMIO) {
+			hw->cfg.p = ioremap(hw->cfg.start, hw->cfg.size);
+			if (!hw->cfg.p)
+				return -ENOMEM;
+		}
 		if (debug & DEBUG_HW)
 			pr_notice("%s: IO cfg %lx (%lu bytes) mode%d\n",
 				  hw->name, (ulong)hw->cfg.start,

@@ -697,9 +702,12 @@ setup_io(struct inf_hw *hw)
 				(ulong)hw->addr.start, (ulong)hw->addr.size);
 			return err;
 		}
-		if (hw->ci->addr_mode == AM_MEMIO)
-			hw->addr.p = ioremap(hw->addr.start, hw->addr.size);
 		hw->addr.mode = hw->ci->addr_mode;
+		if (hw->ci->addr_mode == AM_MEMIO) {
+			hw->addr.p = ioremap(hw->addr.start, hw->addr.size);
+			if (!hw->addr.p)
+				return -ENOMEM;
+		}
 		if (debug & DEBUG_HW)
 			pr_notice("%s: IO addr %lx (%lu bytes) mode%d\n",
 				  hw->name, (ulong)hw->addr.start,

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #103 lore raw

PATCH 103/177 Revert "ath6kl: return error code in ath6kl_wmi_set_roam_lrssi_cmd()"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit efba106f89fc6848726716c101f4c84e88720a9c ]

This reverts commit fc6a6521556c8250e356ddc6a3f2391aa62dc976.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The change being reverted does NOTHING as the caller to this function does not even look at the return value of the call. So the "claim" that this fixed an an issue is not true. It will be fixed up properly in a future patch by propagating the error up the stack correctly.

Cc Kangjie Lu <[email protected]>

Cc Kalle Valo <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/wireless/ath/ath6kl/wmi.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

drivers/net/wireless/ath/ath6kl/wmi.c +3 -1

index c610fe21c85c..d27b4088b874 100644
@@ -776,8 +776,10 @@ int ath6kl_wmi_set_roam_lrssi_cmd(struct wmi *wmi, u8 lrssi)
 	cmd->info.params.roam_rssi_floor = DEF_LRSSI_ROAM_FLOOR;
 	cmd->roam_ctrl = WMI_SET_LRSSI_SCAN_PARAMS;
 
-	return ath6kl_wmi_cmd_send(wmi, 0, skb, WMI_SET_ROAM_CTRL_CMDID,
+	ath6kl_wmi_cmd_send(wmi, 0, skb, WMI_SET_ROAM_CTRL_CMDID,
 			    NO_SYNC_WMIFLAG);
+
+	return 0;
 }
 
 int ath6kl_wmi_force_roam_cmd(struct wmi *wmi, const u8 *bssid)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #104 lore raw

PATCH 104/177 ath6kl: return error code in ath6kl_wmi_set_roam_lrssi_cmd()

From: Anirudh Rayabharam <[email protected]>

[ Upstream commit 54433367840b46a1555c8ed36c4c0cfc5dbf1358 ]

Propagate error code from failure of ath6kl_wmi_cmd_send() to the caller.

Signed-off-by Anirudh Rayabharam <[email protected]>

Cc Kalle Valo <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/wireless/ath/ath6kl/debug.c | 5 ++++-
 drivers/net/wireless/ath/ath6kl/wmi.c   | 4 +---
 2 files changed, 5 insertions(+), 4 deletions(-)

drivers/net/wireless/ath/ath6kl/debug.c +4 -1

index 54337d60f288..085a134069f7 100644
@@ -1027,14 +1027,17 @@ static ssize_t ath6kl_lrssi_roam_write(struct file *file,
 {
 	struct ath6kl *ar = file->private_data;
 	unsigned long lrssi_roam_threshold;
+	int ret;
 
 	if (kstrtoul_from_user(user_buf, count, 0, &lrssi_roam_threshold))
 		return -EINVAL;
 
 	ar->lrssi_roam_threshold = lrssi_roam_threshold;
 
-	ath6kl_wmi_set_roam_lrssi_cmd(ar->wmi, ar->lrssi_roam_threshold);
+	ret = ath6kl_wmi_set_roam_lrssi_cmd(ar->wmi, ar->lrssi_roam_threshold);
 
+	if (ret)
+		return ret;
 	return count;
 }

drivers/net/wireless/ath/ath6kl/wmi.c +1 -3

index d27b4088b874..c610fe21c85c 100644
@@ -776,10 +776,8 @@ int ath6kl_wmi_set_roam_lrssi_cmd(struct wmi *wmi, u8 lrssi)
 	cmd->info.params.roam_rssi_floor = DEF_LRSSI_ROAM_FLOOR;
 	cmd->roam_ctrl = WMI_SET_LRSSI_SCAN_PARAMS;
 
-	ath6kl_wmi_cmd_send(wmi, 0, skb, WMI_SET_ROAM_CTRL_CMDID,
+	return ath6kl_wmi_cmd_send(wmi, 0, skb, WMI_SET_ROAM_CTRL_CMDID,
 			    NO_SYNC_WMIFLAG);
-
-	return 0;
 }
 
 int ath6kl_wmi_force_roam_cmd(struct wmi *wmi, const u8 *bssid)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #105 lore raw

PATCH 105/177 Revert "isdn: mISDN: Fix potential NULL pointer dereference of kzalloc"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 36a2c87f7ed9e305d05b9a5c044cc6c494771504 ]

This reverts commit 38d22659803a033b1b66cd2624c33570c0dde77d.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

While it looks like the original change is correct, it is not, as none of the setup actually happens, and the error value is not propagated upwards.

Cc Aditya Pakki <[email protected]>

Cc David S. Miller <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/isdn/hardware/mISDN/hfcsusb.c | 3 ---
 1 file changed, 3 deletions(-)

drivers/isdn/hardware/mISDN/hfcsusb.c +0 -3

index 008a74a1ed44..7a051435c406 100644
@@ -249,9 +249,6 @@ hfcsusb_ph_info(struct hfcsusb *hw)
 	int i;
 
 	phi = kzalloc(struct_size(phi, bch, dch->dev.nrbchan), GFP_ATOMIC);
-	if (!phi)
-		return;
-
 	phi->dch.ch.protocol = hw->protocol;
 	phi->dch.ch.Flags = dch->Flags;
 	phi->dch.state = dch->state;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #106 lore raw

PATCH 106/177 isdn: mISDN: correctly handle ph_info allocation failure in hfcsusb_ph_info

From: Phillip Potter <[email protected]>

[ Upstream commit 5265db2ccc735e2783b790d6c19fb5cee8c025ed ]

Modify return type of hfcusb_ph_info to int, so that we can pass error value up the call stack when allocation of ph_info fails. Also change three of four call sites to actually account for the memory failure. The fourth, in ph_state_nt, is infeasible to change as it is in turn called by ph_state which is used as a function pointer argument to mISDN_initdchannel, which would necessitate changing its signature and updating all the places where it is used (too many).

Fixes original flawed commit (38d22659803a) from the University of Minnesota.

Cc David S. Miller <[email protected]>

Signed-off-by Phillip Potter <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/isdn/hardware/mISDN/hfcsusb.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

drivers/isdn/hardware/mISDN/hfcsusb.c +10 -8

index 7a051435c406..1f89378b5623 100644
@@ -46,7 +46,7 @@ static void hfcsusb_start_endpoint(struct hfcsusb *hw, int channel);
 static void hfcsusb_stop_endpoint(struct hfcsusb *hw, int channel);
 static int  hfcsusb_setup_bch(struct bchannel *bch, int protocol);
 static void deactivate_bchannel(struct bchannel *bch);
-static void hfcsusb_ph_info(struct hfcsusb *hw);
+static int  hfcsusb_ph_info(struct hfcsusb *hw);
 
 /* start next background transfer for control channel */
 static void

@@ -241,7 +241,7 @@ hfcusb_l2l1B(struct mISDNchannel *ch, struct sk_buff *skb)
  * send full D/B channel status information
  * as MPH_INFORMATION_IND
  */
-static void
+static int
 hfcsusb_ph_info(struct hfcsusb *hw)
 {
 	struct ph_info *phi;

@@ -249,6 +249,9 @@ hfcsusb_ph_info(struct hfcsusb *hw)
 	int i;
 
 	phi = kzalloc(struct_size(phi, bch, dch->dev.nrbchan), GFP_ATOMIC);
+	if (!phi)
+		return -ENOMEM;
+
 	phi->dch.ch.protocol = hw->protocol;
 	phi->dch.ch.Flags = dch->Flags;
 	phi->dch.state = dch->state;

@@ -261,6 +264,8 @@ hfcsusb_ph_info(struct hfcsusb *hw)
 		    sizeof(struct ph_info_dch) + dch->dev.nrbchan *
 		    sizeof(struct ph_info_ch), phi, GFP_ATOMIC);
 	kfree(phi);
+
+	return 0;
 }
 
 /*

@@ -345,8 +350,7 @@ hfcusb_l2l1D(struct mISDNchannel *ch, struct sk_buff *skb)
 			ret = l1_event(dch->l1, hh->prim);
 		break;
 	case MPH_INFORMATION_REQ:
-		hfcsusb_ph_info(hw);
-		ret = 0;
+		ret = hfcsusb_ph_info(hw);
 		break;
 	}

@@ -401,8 +405,7 @@ hfc_l1callback(struct dchannel *dch, u_int cmd)
 			       hw->name, __func__, cmd);
 		return -1;
 	}
-	hfcsusb_ph_info(hw);
-	return 0;
+	return hfcsusb_ph_info(hw);
 }
 
 static int

@@ -744,8 +747,7 @@ hfcsusb_setup_bch(struct bchannel *bch, int protocol)
 			handle_led(hw, (bch->nr == 1) ? LED_B1_OFF :
 				   LED_B2_OFF);
 	}
-	hfcsusb_ph_info(hw);
-	return 0;
+	return hfcsusb_ph_info(hw);
 }
 
 static void

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #107 lore raw

PATCH 107/177 Revert "dmaengine: qcom_hidma: Check for driver register failure"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 43ed0fcf613a87dd0221ec72d1ade4d6544f2ffc ]

This reverts commit a474b3f0428d6b02a538aa10b3c3b722751cb382.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The original change is NOT correct, as it does not correctly unwind from the resources that was allocated before the call to platform_driver_register().

Cc Aditya Pakki <[email protected]>

Acked-by Vinod Koul <[email protected]>

Acked-by Sinan Kaya <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/dma/qcom/hidma_mgmt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/dma/qcom/hidma_mgmt.c +2 -1

index 806ca02c52d7..fe87b01f7a4e 100644
@@ -418,8 +418,9 @@ static int __init hidma_mgmt_init(void)
 		hidma_mgmt_of_populate_channels(child);
 	}
 #endif
-	return platform_driver_register(&hidma_mgmt_driver);
+	platform_driver_register(&hidma_mgmt_driver);
 
+	return 0;
 }
 module_init(hidma_mgmt_init);
 MODULE_LICENSE("GPL v2");

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #108 lore raw

PATCH 108/177 dmaengine: qcom_hidma: comment platform_driver_register call

From: Phillip Potter <[email protected]>

[ Upstream commit 4df2a8b0ad634d98a67e540a4e18a60f943e7d9f ]

Place a comment in hidma_mgmt_init explaining why success must currently be assumed, due to the cleanup issue that would need to be considered were this module ever to be unloadable or were this platform_driver_register call ever to fail.

Acked-by Vinod Koul <[email protected]>

Acked-by Sinan Kaya <[email protected]>

Signed-off-by Phillip Potter <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/dma/qcom/hidma_mgmt.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

drivers/dma/qcom/hidma_mgmt.c +14 -0

index fe87b01f7a4e..62026607f3f8 100644
@@ -418,6 +418,20 @@ static int __init hidma_mgmt_init(void)
 		hidma_mgmt_of_populate_channels(child);
 	}
 #endif
+	/*
+	 * We do not check for return value here, as it is assumed that
+	 * platform_driver_register must not fail. The reason for this is that
+	 * the (potential) hidma_mgmt_of_populate_channels calls above are not
+	 * cleaned up if it does fail, and to do this work is quite
+	 * complicated. In particular, various calls of of_address_to_resource,
+	 * of_irq_to_resource, platform_device_register_full, of_dma_configure,
+	 * and of_msi_configure which then call other functions and so on, must
+	 * be cleaned up - this is not a trivial exercise.
+	 *
+	 * Currently, this module is not intended to be unloaded, and there is
+	 * no module_exit function defined which does the needed cleanup. For
+	 * this reason, we have to assume success here.
+	 */
 	platform_driver_register(&hidma_mgmt_driver);
 
 	return 0;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #109 lore raw

PATCH 109/177 Revert "libertas: add checks for the return value of sysfs_create_group"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 46651077765c80a0d6f87f3469129a72e49ce91b ]

This reverts commit 434256833d8eb988cb7f3b8a41699e2fe48d9332.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The original commit was incorrect, the error needs to be propagated back to the caller AND if the second group call fails, the first needs to be removed. There are much better ways to solve this, the driver should NOT be calling sysfs_create_group() on its own as it is racing userspace and loosing.

Cc Kangjie Lu <[email protected]>

Cc Kalle Valo <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/wireless/marvell/libertas/mesh.c | 5 -----
 1 file changed, 5 deletions(-)

drivers/net/wireless/marvell/libertas/mesh.c +0 -5

index 2747c957d18c..a21c86d446fa 100644
@@ -805,12 +805,7 @@ static void lbs_persist_config_init(struct net_device *dev)
 {
 	int ret;
 	ret = sysfs_create_group(&(dev->dev.kobj), &boot_opts_group);
-	if (ret)
-		pr_err("failed to create boot_opts_group.\n");
-
 	ret = sysfs_create_group(&(dev->dev.kobj), &mesh_ie_group);
-	if (ret)
-		pr_err("failed to create mesh_ie_group.\n");
 }
 
 static void lbs_persist_config_remove(struct net_device *dev)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #110 lore raw

PATCH 110/177 libertas: register sysfs groups properly

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 7e79b38fe9a403b065ac5915465f620a8fb3de84 ]

The libertas driver was trying to register sysfs groups "by hand" which causes them to be created _after_ the device is initialized and announced to userspace, which causes races and can prevent userspace tools from seeing the sysfs files correctly.

Fix this up by using the built-in sysfs_groups pointers in struct net_device which were created for this very reason, fixing the race condition, and properly allowing for any error that might have occured to be handled properly.

Cc Kalle Valo <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/wireless/marvell/libertas/mesh.c | 28 +++-----------------
 1 file changed, 4 insertions(+), 24 deletions(-)

drivers/net/wireless/marvell/libertas/mesh.c +4 ~~-24~~

index a21c86d446fa..050fd403110e 100644
@@ -801,19 +801,6 @@ static const struct attribute_group mesh_ie_group = {
 	.attrs = mesh_ie_attrs,
 };
 
-static void lbs_persist_config_init(struct net_device *dev)
-{
-	int ret;
-	ret = sysfs_create_group(&(dev->dev.kobj), &boot_opts_group);
-	ret = sysfs_create_group(&(dev->dev.kobj), &mesh_ie_group);
-}
-
-static void lbs_persist_config_remove(struct net_device *dev)
-{
-	sysfs_remove_group(&(dev->dev.kobj), &boot_opts_group);
-	sysfs_remove_group(&(dev->dev.kobj), &mesh_ie_group);
-}
-
 
 /***************************************************************************
  * Initializing and starting, stopping mesh

@@ -1014,6 +1001,10 @@ static int lbs_add_mesh(struct lbs_private *priv)
 	SET_NETDEV_DEV(priv->mesh_dev, priv->dev->dev.parent);
 
 	mesh_dev->flags |= IFF_BROADCAST | IFF_MULTICAST;
+	mesh_dev->sysfs_groups[0] = &lbs_mesh_attr_group;
+	mesh_dev->sysfs_groups[1] = &boot_opts_group;
+	mesh_dev->sysfs_groups[2] = &mesh_ie_group;
+
 	/* Register virtual mesh interface */
 	ret = register_netdev(mesh_dev);
 	if (ret) {

@@ -1021,19 +1012,10 @@ static int lbs_add_mesh(struct lbs_private *priv)
 		goto err_free_netdev;
 	}
 
-	ret = sysfs_create_group(&(mesh_dev->dev.kobj), &lbs_mesh_attr_group);
-	if (ret)
-		goto err_unregister;
-
-	lbs_persist_config_init(mesh_dev);
-
 	/* Everything successful */
 	ret = 0;
 	goto done;
 
-err_unregister:
-	unregister_netdev(mesh_dev);
-
 err_free_netdev:
 	free_netdev(mesh_dev);

@@ -1054,8 +1036,6 @@ void lbs_remove_mesh(struct lbs_private *priv)
 
 	netif_stop_queue(mesh_dev);
 	netif_carrier_off(mesh_dev);
-	sysfs_remove_group(&(mesh_dev->dev.kobj), &lbs_mesh_attr_group);
-	lbs_persist_config_remove(mesh_dev);
 	unregister_netdev(mesh_dev);
 	priv->mesh_dev = NULL;
 	kfree(mesh_dev->ieee80211_ptr);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #111 lore raw

PATCH 111/177 Revert "ASoC: cs43130: fix a NULL pointer dereference"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit fdda0dd2686ecd1f2e616c9e0366ea71b40c485d ]

This reverts commit a2be42f18d409213bb7e7a736e3ef6ba005115bb.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The original patch here is not correct, sysfs files that were created are not unwound.

Cc Kangjie Lu <[email protected]>

Cc Mark Brown <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 sound/soc/codecs/cs43130.c | 2 --
 1 file changed, 2 deletions(-)

sound/soc/codecs/cs43130.c +0 -2

index 7fb34422a2a4..bb46e993c353 100644
@@ -2319,8 +2319,6 @@ static int cs43130_probe(struct snd_soc_component *component)
 			return ret;
 
 		cs43130->wq = create_singlethread_workqueue("cs43130_hp");
-		if (!cs43130->wq)
-			return -ENOMEM;
 		INIT_WORK(&cs43130->work, cs43130_imp_meas);
 	}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #112 lore raw

PATCH 112/177 ASoC: cs43130: handle errors in cs43130_probe() properly

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 2da441a6491d93eff8ffff523837fd621dc80389 ]

cs43130_probe() does not do any valid error checking of things it initializes, OR what it does, it does not unwind properly if there are errors.

Fix this up by moving the sysfs files to an attribute group so the driver core will correctly add/remove them all at once and handle errors with them, and correctly check for creating a new workqueue and unwinding if that fails.

Cc Mark Brown <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 sound/soc/codecs/cs43130.c | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

sound/soc/codecs/cs43130.c +14 ~~-14~~

index bb46e993c353..8f70dee95878 100644
@@ -1735,6 +1735,14 @@ static DEVICE_ATTR(hpload_dc_r, 0444, cs43130_show_dc_r, NULL);
 static DEVICE_ATTR(hpload_ac_l, 0444, cs43130_show_ac_l, NULL);
 static DEVICE_ATTR(hpload_ac_r, 0444, cs43130_show_ac_r, NULL);
 
+static struct attribute *hpload_attrs[] = {
+	&dev_attr_hpload_dc_l.attr,
+	&dev_attr_hpload_dc_r.attr,
+	&dev_attr_hpload_ac_l.attr,
+	&dev_attr_hpload_ac_r.attr,
+};
+ATTRIBUTE_GROUPS(hpload);
+
 static struct reg_sequence hp_en_cal_seq[] = {
 	{CS43130_INT_MASK_4, CS43130_INT_MASK_ALL},
 	{CS43130_HP_MEAS_LOAD_1, 0},

@@ -2302,23 +2310,15 @@ static int cs43130_probe(struct snd_soc_component *component)
 
 	cs43130->hpload_done = false;
 	if (cs43130->dc_meas) {
-		ret = device_create_file(component->dev, &dev_attr_hpload_dc_l);
-		if (ret < 0)
-			return ret;
-
-		ret = device_create_file(component->dev, &dev_attr_hpload_dc_r);
-		if (ret < 0)
-			return ret;
-
-		ret = device_create_file(component->dev, &dev_attr_hpload_ac_l);
-		if (ret < 0)
-			return ret;
-
-		ret = device_create_file(component->dev, &dev_attr_hpload_ac_r);
-		if (ret < 0)
+		ret = sysfs_create_groups(&component->dev->kobj, hpload_groups);
+		if (ret)
 			return ret;
 
 		cs43130->wq = create_singlethread_workqueue("cs43130_hp");
+		if (!cs43130->wq) {
+			sysfs_remove_groups(&component->dev->kobj, hpload_groups);
+			return -ENOMEM;
+		}
 		INIT_WORK(&cs43130->work, cs43130_imp_meas);
 	}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:23 UTC #113 lore raw

PATCH 113/177 Revert "media: dvb: Add check on sp8870_readreg"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 47e4ff06fa7f5ba4860543a2913bbd0c164640aa ]

This reverts commit 467a37fba93f2b4fe3ab597ff6a517b22b566882.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

This commit is not properly checking for an error at all, so if a read succeeds from this device, it will error out.

Cc Aditya Pakki <[email protected]>

Cc Sean Young <[email protected]>

Cc Mauro Carvalho Chehab <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/media/dvb-frontends/sp8870.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

drivers/media/dvb-frontends/sp8870.c +1 -3

index 655db8272268..ee893a2f2261 100644
@@ -280,9 +280,7 @@ static int sp8870_set_frontend_parameters(struct dvb_frontend *fe)
 	sp8870_writereg(state, 0xc05, reg0xc05);
 
 	// read status reg in order to clear pending irqs
-	err = sp8870_readreg(state, 0x200);
-	if (err)
-		return err;
+	sp8870_readreg(state, 0x200);
 
 	// system controller start
 	sp8870_microcontroller_start(state);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:24 UTC #114 lore raw

PATCH 114/177 media: dvb: Add check on sp8870_readreg return

From: Alaa Emad <[email protected]>

[ Upstream commit c6d822c56e7fd29e6fa1b1bb91b98f6a1e942b3c ]

The function sp8870_readreg returns a negative value when i2c_transfer fails so properly check for this and return the error if it happens.

Cc Sean Young <[email protected]>

Cc Mauro Carvalho Chehab <[email protected]>

Signed-off-by Alaa Emad <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/media/dvb-frontends/sp8870.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

drivers/media/dvb-frontends/sp8870.c +3 -1

index ee893a2f2261..9767159aeb9b 100644
@@ -280,7 +280,9 @@ static int sp8870_set_frontend_parameters(struct dvb_frontend *fe)
 	sp8870_writereg(state, 0xc05, reg0xc05);
 
 	// read status reg in order to clear pending irqs
-	sp8870_readreg(state, 0x200);
+	err = sp8870_readreg(state, 0x200);
+	if (err < 0)
+		return err;
 
 	// system controller start
 	sp8870_microcontroller_start(state);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:24 UTC #115 lore raw

PATCH 115/177 Revert "media: gspca: mt9m111: Check write_bridge for timeout"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit d8c3be2fb2079d0cb4cd29d6aba58dbe54771e42 ]

This reverts commit 656025850074f5c1ba2e05be37bda57ba2b8d491.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

Different error values should never be "OR" together and expect anything sane to come out of the result.

Cc Aditya Pakki <[email protected]>

Cc Mauro Carvalho Chehab <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/media/usb/gspca/m5602/m5602_mt9m111.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

drivers/media/usb/gspca/m5602/m5602_mt9m111.c +3 -5

index bfa3b381d8a2..50481dc928d0 100644
@@ -195,7 +195,7 @@ static const struct v4l2_ctrl_config mt9m111_greenbal_cfg = {
 int mt9m111_probe(struct sd *sd)
 {
 	u8 data[2] = {0x00, 0x00};
-	int i, rc = 0;
+	int i;
 	struct gspca_dev *gspca_dev = (struct gspca_dev *)sd;
 
 	if (force_sensor) {

@@ -213,18 +213,16 @@ int mt9m111_probe(struct sd *sd)
 	/* Do the preinit */
 	for (i = 0; i < ARRAY_SIZE(preinit_mt9m111); i++) {
 		if (preinit_mt9m111[i][0] == BRIDGE) {
-			rc |= m5602_write_bridge(sd,
+			m5602_write_bridge(sd,
 				preinit_mt9m111[i][1],
 				preinit_mt9m111[i][2]);
 		} else {
 			data[0] = preinit_mt9m111[i][2];
 			data[1] = preinit_mt9m111[i][3];
-			rc |= m5602_write_sensor(sd,
+			m5602_write_sensor(sd,
 				preinit_mt9m111[i][1], data, 2);
 		}
 	}
-	if (rc < 0)
-		return rc;
 
 	if (m5602_read_sensor(sd, MT9M111_SC_CHIPVER, data, 2))
 		return -ENODEV;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:24 UTC #116 lore raw

PATCH 116/177 media: gspca: mt9m111: Check write_bridge for timeout

From: Alaa Emad <[email protected]>

[ Upstream commit e932f5b458eee63d013578ea128b9ff8ef5f5496 ]

If m5602_write_bridge times out, it will return a negative error value. So properly check for this and handle the error correctly instead of just ignoring it.

Cc Mauro Carvalho Chehab <[email protected]>

Signed-off-by Alaa Emad <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/media/usb/gspca/m5602/m5602_mt9m111.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

drivers/media/usb/gspca/m5602/m5602_mt9m111.c +8 -6

index 50481dc928d0..bf1af6ed9131 100644
@@ -195,7 +195,7 @@ static const struct v4l2_ctrl_config mt9m111_greenbal_cfg = {
 int mt9m111_probe(struct sd *sd)
 {
 	u8 data[2] = {0x00, 0x00};
-	int i;
+	int i, err;
 	struct gspca_dev *gspca_dev = (struct gspca_dev *)sd;
 
 	if (force_sensor) {

@@ -213,15 +213,17 @@ int mt9m111_probe(struct sd *sd)
 	/* Do the preinit */
 	for (i = 0; i < ARRAY_SIZE(preinit_mt9m111); i++) {
 		if (preinit_mt9m111[i][0] == BRIDGE) {
-			m5602_write_bridge(sd,
-				preinit_mt9m111[i][1],
-				preinit_mt9m111[i][2]);
+			err = m5602_write_bridge(sd,
+					preinit_mt9m111[i][1],
+					preinit_mt9m111[i][2]);
 		} else {
 			data[0] = preinit_mt9m111[i][2];
 			data[1] = preinit_mt9m111[i][3];
-			m5602_write_sensor(sd,
-				preinit_mt9m111[i][1], data, 2);
+			err = m5602_write_sensor(sd,
+					preinit_mt9m111[i][1], data, 2);
 		}
+		if (err < 0)
+			return err;
 	}
 
 	if (m5602_read_sensor(sd, MT9M111_SC_CHIPVER, data, 2))

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:24 UTC #117 lore raw

PATCH 117/177 Revert "media: gspca: Check the return value of write_bridge for timeout"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 8e23e83c752b54e98102627a1cc09281ad71a299 ]

This reverts commit a21a0eb56b4e8fe4a330243af8030f890cde2283.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

Different error values should never be "OR" together and expect anything sane to come out of the result.

Cc Aditya Pakki <[email protected]>

Cc Mauro Carvalho Chehab <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/media/usb/gspca/m5602/m5602_po1030.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

drivers/media/usb/gspca/m5602/m5602_po1030.c +2 -6

index d680b777f097..7bdbb8065146 100644
@@ -154,7 +154,6 @@ static const struct v4l2_ctrl_config po1030_greenbal_cfg = {
 
 int po1030_probe(struct sd *sd)
 {
-	int rc = 0;
 	u8 dev_id_h = 0, i;
 	struct gspca_dev *gspca_dev = (struct gspca_dev *)sd;

@@ -174,14 +173,11 @@ int po1030_probe(struct sd *sd)
 	for (i = 0; i < ARRAY_SIZE(preinit_po1030); i++) {
 		u8 data = preinit_po1030[i][2];
 		if (preinit_po1030[i][0] == SENSOR)
-			rc |= m5602_write_sensor(sd,
+			m5602_write_sensor(sd,
 				preinit_po1030[i][1], &data, 1);
 		else
-			rc |= m5602_write_bridge(sd, preinit_po1030[i][1],
-						data);
+			m5602_write_bridge(sd, preinit_po1030[i][1], data);
 	}
-	if (rc < 0)
-		return rc;
 
 	if (m5602_read_sensor(sd, PO1030_DEVID_H, &dev_id_h, 1))
 		return -ENODEV;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:24 UTC #118 lore raw

PATCH 118/177 media: gspca: properly check for errors in po1030_probe()

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit dacb408ca6f0e34df22b40d8dd5fae7f8e777d84 ]

If m5602_write_sensor() or m5602_write_bridge() fail, do not continue to initialize the device but return the error to the calling funtion.

Cc Mauro Carvalho Chehab <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/media/usb/gspca/m5602/m5602_po1030.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

drivers/media/usb/gspca/m5602/m5602_po1030.c +7 -3

index 7bdbb8065146..8fd99ceee4b6 100644
@@ -155,6 +155,7 @@ static const struct v4l2_ctrl_config po1030_greenbal_cfg = {
 int po1030_probe(struct sd *sd)
 {
 	u8 dev_id_h = 0, i;
+	int err;
 	struct gspca_dev *gspca_dev = (struct gspca_dev *)sd;
 
 	if (force_sensor) {

@@ -173,10 +174,13 @@ int po1030_probe(struct sd *sd)
 	for (i = 0; i < ARRAY_SIZE(preinit_po1030); i++) {
 		u8 data = preinit_po1030[i][2];
 		if (preinit_po1030[i][0] == SENSOR)
-			m5602_write_sensor(sd,
-				preinit_po1030[i][1], &data, 1);
+			err = m5602_write_sensor(sd, preinit_po1030[i][1],
+						 &data, 1);
 		else
-			m5602_write_bridge(sd, preinit_po1030[i][1], data);
+			err = m5602_write_bridge(sd, preinit_po1030[i][1],
+						 data);
+		if (err < 0)
+			return err;
 	}
 
 	if (m5602_read_sensor(sd, PO1030_DEVID_H, &dev_id_h, 1))

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:24 UTC #119 lore raw

PATCH 119/177 Revert "net: liquidio: fix a NULL pointer dereference"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 4fd798a5a89114c1892574c50f2aebd49bc5b4f5 ]

This reverts commit fe543b2f174f34a7a751aa08b334fe6b105c4569.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

While the original commit does keep the immediate "NULL dereference" from happening, it does not properly propagate the error back to the callers, AND it does not fix this same identical issue in the drivers/net/ethernet/cavium/liquidio/lio_vf_main.c for some reason.

Cc Kangjie Lu <[email protected]>

Cc David S. Miller <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/cavium/liquidio/lio_main.c | 5 -----
 1 file changed, 5 deletions(-)

drivers/net/ethernet/cavium/liquidio/lio_main.c +0 -5

index 7f3b2e3b0868..f2d486583e2f 100644
@@ -1192,11 +1192,6 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 	sc = (struct octeon_soft_command *)
 		octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
 					  16, 0);
-	if (!sc) {
-		netif_info(lio, rx_err, lio->netdev,
-			   "Failed to allocate octeon_soft_command\n");
-		return;
-	}
 
 	ncmd = (union octnet_cmd *)sc->virtdptr;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:24 UTC #120 lore raw

PATCH 120/177 net: liquidio: Add missing null pointer checks

From: Tom Seewald <[email protected]>

[ Upstream commit dbc97bfd3918ed9268bfc174cae8a7d6b3d51aad ]

The functions send_rx_ctrl_cmd() in both liquidio/lio_main.c and liquidio/lio_vf_main.c do not check if the call to octeon_alloc_soft_command() fails and returns a null pointer. Both functions also return void so errors are not propagated back to the caller.

Fix these issues by updating both instances of send_rx_ctrl_cmd() to return an integer rather than void, and have them return -ENOMEM if an allocation failure occurs. Also update all callers of send_rx_ctrl_cmd() so that they now check the return value.

Cc David S. Miller <[email protected]>

Signed-off-by Tom Seewald <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 .../net/ethernet/cavium/liquidio/lio_main.c   | 28 +++++++++++++------
 .../ethernet/cavium/liquidio/lio_vf_main.c    | 27 +++++++++++++-----
 2 files changed, 40 insertions(+), 15 deletions(-)

drivers/net/ethernet/cavium/liquidio/lio_main.c +20 -8

index f2d486583e2f..d0c77ff9dbb1 100644
@@ -1179,7 +1179,7 @@ static void octeon_destroy_resources(struct octeon_device *oct)
  * @param lio per-network private data
  * @param start_stop whether to start or stop
  */
-static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
+static int send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 {
 	struct octeon_soft_command *sc;
 	union octnet_cmd *ncmd;

@@ -1187,11 +1187,16 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 	int retval;
 
 	if (oct->props[lio->ifidx].rx_on == start_stop)
-		return;
+		return 0;
 
 	sc = (struct octeon_soft_command *)
 		octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
 					  16, 0);
+	if (!sc) {
+		netif_info(lio, rx_err, lio->netdev,
+			   "Failed to allocate octeon_soft_command struct\n");
+		return -ENOMEM;
+	}
 
 	ncmd = (union octnet_cmd *)sc->virtdptr;

@@ -1213,18 +1218,19 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 	if (retval == IQ_SEND_FAILED) {
 		netif_info(lio, rx_err, lio->netdev, "Failed to send RX Control message\n");
 		octeon_free_soft_command(oct, sc);
-		return;
 	} else {
 		/* Sleep on a wait queue till the cond flag indicates that the
 		 * response arrived or timed-out.
 		 */
 		retval = wait_for_sc_completion_timeout(oct, sc, 0);
 		if (retval)
-			return;
+			return retval;
 
 		oct->props[lio->ifidx].rx_on = start_stop;
 		WRITE_ONCE(sc->caller_is_done, true);
 	}
+
+	return retval;
 }
 
 /**

@@ -1811,6 +1817,7 @@ static int liquidio_open(struct net_device *netdev)
 	struct octeon_device_priv *oct_priv =
 		(struct octeon_device_priv *)oct->priv;
 	struct napi_struct *napi, *n;
+	int ret = 0;
 
 	if (oct->props[lio->ifidx].napi_enabled == 0) {
 		tasklet_disable(&oct_priv->droq_tasklet);

@@ -1846,7 +1853,9 @@ static int liquidio_open(struct net_device *netdev)
 	netif_info(lio, ifup, lio->netdev, "Interface Open, ready for traffic\n");
 
 	/* tell Octeon to start forwarding packets to host */
-	send_rx_ctrl_cmd(lio, 1);
+	ret = send_rx_ctrl_cmd(lio, 1);
+	if (ret)
+		return ret;
 
 	/* start periodical statistics fetch */
 	INIT_DELAYED_WORK(&lio->stats_wk.work, lio_fetch_stats);

@@ -1857,7 +1866,7 @@ static int liquidio_open(struct net_device *netdev)
 	dev_info(&oct->pci_dev->dev, "%s interface is opened\n",
 		 netdev->name);
 
-	return 0;
+	return ret;
 }
 
 /**

@@ -1871,6 +1880,7 @@ static int liquidio_stop(struct net_device *netdev)
 	struct octeon_device_priv *oct_priv =
 		(struct octeon_device_priv *)oct->priv;
 	struct napi_struct *napi, *n;
+	int ret = 0;
 
 	ifstate_reset(lio, LIO_IFSTATE_RUNNING);

@@ -1887,7 +1897,9 @@ static int liquidio_stop(struct net_device *netdev)
 	lio->link_changes++;
 
 	/* Tell Octeon that nic interface is down. */
-	send_rx_ctrl_cmd(lio, 0);
+	ret = send_rx_ctrl_cmd(lio, 0);
+	if (ret)
+		return ret;
 
 	if (OCTEON_CN23XX_PF(oct)) {
 		if (!oct->msix_on)

@@ -1922,7 +1934,7 @@ static int liquidio_stop(struct net_device *netdev)
 
 	dev_info(&oct->pci_dev->dev, "%s interface is stopped\n", netdev->name);
 
-	return 0;
+	return ret;
 }
 
 /**

drivers/net/ethernet/cavium/liquidio/lio_vf_main.c +20 -7

index 370d76822ee0..929da9e9fe9a 100644
@@ -598,7 +598,7 @@ static void octeon_destroy_resources(struct octeon_device *oct)
  * @param lio per-network private data
  * @param start_stop whether to start or stop
  */
-static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
+static int send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 {
 	struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;
 	struct octeon_soft_command *sc;

@@ -606,11 +606,16 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 	int retval;
 
 	if (oct->props[lio->ifidx].rx_on == start_stop)
-		return;
+		return 0;
 
 	sc = (struct octeon_soft_command *)
 		octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
 					  16, 0);
+	if (!sc) {
+		netif_info(lio, rx_err, lio->netdev,
+			   "Failed to allocate octeon_soft_command struct\n");
+		return -ENOMEM;
+	}
 
 	ncmd = (union octnet_cmd *)sc->virtdptr;

@@ -638,11 +643,13 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop)
 		 */
 		retval = wait_for_sc_completion_timeout(oct, sc, 0);
 		if (retval)
-			return;
+			return retval;
 
 		oct->props[lio->ifidx].rx_on = start_stop;
 		WRITE_ONCE(sc->caller_is_done, true);
 	}
+
+	return retval;
 }
 
 /**

@@ -909,6 +916,7 @@ static int liquidio_open(struct net_device *netdev)
 	struct octeon_device_priv *oct_priv =
 		(struct octeon_device_priv *)oct->priv;
 	struct napi_struct *napi, *n;
+	int ret = 0;
 
 	if (!oct->props[lio->ifidx].napi_enabled) {
 		tasklet_disable(&oct_priv->droq_tasklet);

@@ -935,11 +943,13 @@ static int liquidio_open(struct net_device *netdev)
 					(LIQUIDIO_NDEV_STATS_POLL_TIME_MS));
 
 	/* tell Octeon to start forwarding packets to host */
-	send_rx_ctrl_cmd(lio, 1);
+	ret = send_rx_ctrl_cmd(lio, 1);
+	if (ret)
+		return ret;
 
 	dev_info(&oct->pci_dev->dev, "%s interface is opened\n", netdev->name);
 
-	return 0;
+	return ret;
 }
 
 /**

@@ -953,9 +963,12 @@ static int liquidio_stop(struct net_device *netdev)
 	struct octeon_device_priv *oct_priv =
 		(struct octeon_device_priv *)oct->priv;
 	struct napi_struct *napi, *n;
+	int ret = 0;
 
 	/* tell Octeon to stop forwarding packets to host */
-	send_rx_ctrl_cmd(lio, 0);
+	ret = send_rx_ctrl_cmd(lio, 0);
+	if (ret)
+		return ret;
 
 	netif_info(lio, ifdown, lio->netdev, "Stopping interface!\n");
 	/* Inform that netif carrier is down */

@@ -989,7 +1002,7 @@ static int liquidio_stop(struct net_device *netdev)
 
 	dev_info(&oct->pci_dev->dev, "%s interface is stopped\n", netdev->name);
 
-	return 0;
+	return ret;
 }
 
 /**

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #121 lore raw

PATCH 121/177 Revert "brcmfmac: add a check for the status of usb_register"

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 30a350947692f794796f563029d29764497f2887 ]

This reverts commit 42daad3343be4a4e1ee03e30a5f5cc731dadfef5.

Because of recent interactions with developers from @umn.edu, all commits from them have been recently re-reviewed to ensure if they were correct or not.

Upon review, this commit was found to be incorrect for the reasons below, so it must be reverted. It will be fixed up "correctly" in a later kernel change.

The original commit here did nothing to actually help if usb_register() failed, so it gives a "false sense of security" when there is none. The correct solution is to correctly unwind from this error.

Cc Kangjie Lu <[email protected]>

Cc Kalle Valo <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c +1 -5

index 575ed19e9195..6f41d28930e4 100644
@@ -1560,10 +1560,6 @@ void brcmf_usb_exit(void)
 
 void brcmf_usb_register(void)
 {
-	int ret;
-
 	brcmf_dbg(USB, "Enter\n");
-	ret = usb_register(&brcmf_usbdrvr);
-	if (ret)
-		brcmf_err("usb_register failed %d\n", ret);
+	usb_register(&brcmf_usbdrvr);
 }

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #122 lore raw

PATCH 122/177 brcmfmac: properly check for bus register errors

From: Greg Kroah-Hartman <[email protected]>

[ Upstream commit 419b4a142a7ece36cebcd434f8ce2af59ef94b85 ]

The brcmfmac driver ignores any errors on initialization with the different busses by deferring the initialization to a workqueue and ignoring all possible errors that might happen. Fix up all of this by only allowing the module to load if all bus registering worked properly.

Cc Kalle Valo <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 .../broadcom/brcm80211/brcmfmac/bcmsdh.c      |  8 +---
 .../broadcom/brcm80211/brcmfmac/bus.h         | 19 ++++++++-
 .../broadcom/brcm80211/brcmfmac/core.c        | 42 ++++++++-----------
 .../broadcom/brcm80211/brcmfmac/pcie.c        |  9 +---
 .../broadcom/brcm80211/brcmfmac/pcie.h        |  5 ---
 .../broadcom/brcm80211/brcmfmac/usb.c         |  4 +-
 6 files changed, 41 insertions(+), 46 deletions(-)

drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c +2 -6

index fc12598b2dd3..c492d2d2db1d 100644
@@ -1168,13 +1168,9 @@ static struct sdio_driver brcmf_sdmmc_driver = {
 	},
 };
 
-void brcmf_sdio_register(void)
+int brcmf_sdio_register(void)
 {
-	int ret;
-
-	ret = sdio_register_driver(&brcmf_sdmmc_driver);
-	if (ret)
-		brcmf_err("sdio_register_driver failed: %d\n", ret);
+	return sdio_register_driver(&brcmf_sdmmc_driver);
 }
 
 void brcmf_sdio_exit(void)

drivers/net/wireless/broadcom/brcm80211/brcmfmac/bus.h +17 -2

index 623c0168da79..8b27494a5d3d 100644
@@ -274,11 +274,26 @@ void brcmf_bus_add_txhdrlen(struct device *dev, uint len);
 
 #ifdef CONFIG_BRCMFMAC_SDIO
 void brcmf_sdio_exit(void);
-void brcmf_sdio_register(void);
+int brcmf_sdio_register(void);
+#else
+static inline void brcmf_sdio_exit(void) { }
+static inline int brcmf_sdio_register(void) { return 0; }
 #endif
+
 #ifdef CONFIG_BRCMFMAC_USB
 void brcmf_usb_exit(void);
-void brcmf_usb_register(void);
+int brcmf_usb_register(void);
+#else
+static inline void brcmf_usb_exit(void) { }
+static inline int brcmf_usb_register(void) { return 0; }
+#endif
+
+#ifdef CONFIG_BRCMFMAC_PCIE
+void brcmf_pcie_exit(void);
+int brcmf_pcie_register(void);
+#else
+static inline void brcmf_pcie_exit(void) { }
+static inline int brcmf_pcie_register(void) { return 0; }
 #endif
 
 #endif /* BRCMFMAC_BUS_H */

drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c +18 ~~-24~~

58 lines not on this page · raw message

drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c +2 -7

index cb68f54a9c56..bda042138e96 100644
@@ -2137,15 +2137,10 @@ static struct pci_driver brcmf_pciedrvr = {
 };
 
 
-void brcmf_pcie_register(void)
+int brcmf_pcie_register(void)
 {
-	int err;
-
 	brcmf_dbg(PCIE, "Enter\n");
-	err = pci_register_driver(&brcmf_pciedrvr);
-	if (err)
-		brcmf_err(NULL, "PCIE driver registration failed, err=%d\n",
-			  err);
+	return pci_register_driver(&brcmf_pciedrvr);
 }

drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.h +0 -5

index d026401d2001..8e6c227e8315 100644
@@ -11,9 +11,4 @@ struct brcmf_pciedev {
 	struct brcmf_pciedev_info *devinfo;
 };
 
-
-void brcmf_pcie_exit(void);
-void brcmf_pcie_register(void);
-
-
 #endif /* BRCMFMAC_PCIE_H */

drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c +2 -2

index 6f41d28930e4..3b897f040371 100644
@@ -1558,8 +1558,8 @@ void brcmf_usb_exit(void)
 	usb_deregister(&brcmf_usbdrvr);
 }
 
-void brcmf_usb_register(void)
+int brcmf_usb_register(void)
 {
 	brcmf_dbg(USB, "Enter\n");
-	usb_register(&brcmf_usbdrvr);
+	return usb_register(&brcmf_usbdrvr);
 }

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #123 lore raw

PATCH 123/177 btrfs: return whole extents in fiemap

From: Boris Burkov <[email protected]>

[ Upstream commit 15c7745c9a0078edad1f7df5a6bb7b80bc8cca23 ]

xfs_io -c 'fiemap <off> <len>' <file>

can give surprising results on btrfs that differ from xfs.

btrfs prints out extents trimmed to fit the user input. If the user's fiemap request has an offset, then rather than returning each whole extent which intersects that range, we also trim the start extent to not have start < off.

Documentation in filesystems/fiemap.txt and the xfs_io man page suggests that returning the whole extent is expected.

Some cases which all yield the same fiemap in xfs, but not btrfs:
dd if=/dev/zero of=$f bs=4k count=1
sudo xfs_io -c 'fiemap 0 1024' $f

    0: [0..7]: 26624..26631
  sudo xfs_io -c 'fiemap 2048 1024' $f
    0: [4..7]: 26628..26631
  sudo xfs_io -c 'fiemap 2048 4096' $f
    0: [4..7]: 26628..26631
  sudo xfs_io -c 'fiemap 3584 512' $f
    0: [7..7]: 26631..26631
  sudo xfs_io -c 'fiemap 4091 5' $f
    0: [7..6]: 26631..26630

I believe this is a consequence of the logic for merging contiguous extents represented by separate extent items. That logic needs to track the last offset as it loops through the extent items, which happens to pick up the start offset on the first iteration, and trim off the beginning of the full extent. To fix it, start off at 0 rather than start so that we keep the iteration/merging intact without cutting off the start of the extent.

after the fix, all the above commands give:

0: [0..7]: 26624..26631

The merging logic is exercised by fstest generic/483, and I have written a new fstest for checking we don't have backwards or zero-length fiemaps for cases like those above.

Reviewed-by Josef Bacik <[email protected]>

Signed-off-by Boris Burkov <[email protected]>

Signed-off-by David Sterba <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 fs/btrfs/extent_io.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

fs/btrfs/extent_io.c +6 -1

index 95205bde240f..eca3abc1a7cd 100644
@@ -4648,7 +4648,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		__u64 start, __u64 len)
 {
 	int ret = 0;
-	u64 off = start;
+	u64 off;
 	u64 max = start + len;
 	u32 flags = 0;
 	u32 found_type;

@@ -4684,6 +4684,11 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		goto out_free_ulist;
 	}
 
+	/*
+	 * We can't initialize that to 'start' as this could miss extents due
+	 * to extent item merging
+	 */
+	off = 0;
 	start = round_down(start, btrfs_inode_sectorsize(inode));
 	len = round_up(max, btrfs_inode_sectorsize(inode)) - start;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:26 UTC #124 lore raw

PATCH 124/177 scsi: BusLogic: Fix 64-bit system enumeration error for Buslogic

From: Matt Wang <[email protected]>

[ Upstream commit 56f396146af278135c0ff958c79b5ee1bd22453d ]

Commit 391e2f25601e ("[SCSI] BusLogic: Port driver to 64-bit") introduced a serious issue for 64-bit systems. With this commit, 64-bit kernel will enumerate 8*15 non-existing disks. This is caused by the broken CCB structure. The change from u32 data to void *data increased CCB length on 64-bit system, which introduced an extra 4 byte offset of the CDB. This leads to incorrect response to INQUIRY commands during enumeration.

Fix disk enumeration failure by reverting the portion of the commit above which switched the data pointer from u32 to void.

Link https://lore.kernel.org/r/[email protected] open here

Acked-by Khalid Aziz <[email protected]>

Signed-off-by Matt Wang <[email protected]>

Signed-off-by Martin K. Petersen <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/scsi/BusLogic.c | 6 +++---
 drivers/scsi/BusLogic.h | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

drivers/scsi/BusLogic.c +3 -3

index c25e8a54e869..6e988233fb81 100644
@@ -3077,11 +3077,11 @@ static int blogic_qcmd_lck(struct scsi_cmnd *command,
 		ccb->opcode = BLOGIC_INITIATOR_CCB_SG;
 		ccb->datalen = count * sizeof(struct blogic_sg_seg);
 		if (blogic_multimaster_type(adapter))
-			ccb->data = (void *)((unsigned int) ccb->dma_handle +
+			ccb->data = (unsigned int) ccb->dma_handle +
 					((unsigned long) &ccb->sglist -
-					(unsigned long) ccb));
+					(unsigned long) ccb);
 		else
-			ccb->data = ccb->sglist;
+			ccb->data = virt_to_32bit_virt(ccb->sglist);
 
 		scsi_for_each_sg(command, sg, count, i) {
 			ccb->sglist[i].segbytes = sg_dma_len(sg);

drivers/scsi/BusLogic.h +1 -1

index 6182cc8a0344..e081ad47d1cf 100644
@@ -814,7 +814,7 @@ struct blogic_ccb {
 	unsigned char cdblen;				/* Byte 2 */
 	unsigned char sense_datalen;			/* Byte 3 */
 	u32 datalen;					/* Bytes 4-7 */
-	void *data;					/* Bytes 8-11 */
+	u32 data;					/* Bytes 8-11 */
 	unsigned char:8;				/* Byte 12 */
 	unsigned char:8;				/* Byte 13 */
 	enum blogic_adapter_status adapter_status;	/* Byte 14 */

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #125 lore raw

PATCH 125/177 openrisc: Define memory barrier mb

From: Peter Zijlstra <[email protected]>

[ Upstream commit 8b549c18ae81dbc36fb11e4aa08b8378c599ca95 ]

This came up in the discussion of the requirements of qspinlock on an architecture. OpenRISC uses qspinlock, but it was noticed that the memmory barrier was not defined.

Peter defined it in the mail thread writing:

As near as I can tell this should do. The arch spec only lists this one instruction and the text makes it sound like a completion barrier.

This is correct so applying this patch.

Signed-off-by Peter Zijlstra <[email protected]>

[[email protected]:Turned the mail into a patch]

Signed-off-by Stafford Horne <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/openrisc/include/asm/barrier.h | 9 +++++++++
 1 file changed, 9 insertions(+)
 create mode 100644 arch/openrisc/include/asm/barrier.h

arch/openrisc/include/asm/barrier.h new file +9 -0

new file mode 100644
index 000000000000..7538294721be
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ASM_BARRIER_H
+#define __ASM_BARRIER_H
+
+#define mb() asm volatile ("l.msync" ::: "memory")
+
+#include <asm-generic/barrier.h>
+
+#endif /* __ASM_BARRIER_H */

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #126 lore raw

PATCH 126/177 btrfs: do not BUG_ON in link_to_fixup_dir

From: Josef Bacik <[email protected]>

[ Upstream commit 91df99a6eb50d5a1bc70fff4a09a0b7ae6aab96d ]

While doing error injection testing I got the following panic

kernel BUG at fs/btrfs/tree-log.c:1862!
invalid opcode: 0000 [#1] SMP NOPTI

  CPU: 1 PID: 7836 Comm: mount Not tainted 5.13.0-rc1+ #305
  Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.13.0-2.fc32 04/01/2014
  RIP: 0010:link_to_fixup_dir+0xd5/0xe0
  RSP: 0018:ffffb5800180fa30 EFLAGS: 00010216
  RAX: fffffffffffffffb RBX: 00000000fffffffb RCX: ffff8f595287faf0
  RDX: ffffb5800180fa37 RSI: ffff8f5954978800 RDI: 0000000000000000
  RBP: ffff8f5953af9450 R08: 0000000000000019 R09: 0000000000000001
  R10: 000151f408682970 R11: 0000000120021001 R12: ffff8f5954978800
  R13: ffff8f595287faf0 R14: ffff8f5953c77dd0 R15: 0000000000000065
  FS:  00007fc5284c8c40(0000) GS:ffff8f59bbd00000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 00007fc5287f47c0 CR3: 000000011275e002 CR4: 0000000000370ee0
  Call Trace:
   replay_one_buffer+0x409/0x470
   ? btree_read_extent_buffer_pages+0xd0/0x110
   walk_up_log_tree+0x157/0x1e0
   walk_log_tree+0xa6/0x1d0
   btrfs_recover_log_trees+0x1da/0x360
   ? replay_one_extent+0x7b0/0x7b0
   open_ctree+0x1486/0x1720
   btrfs_mount_root.cold+0x12/0xea
   ? __kmalloc_track_caller+0x12f/0x240
   legacy_get_tree+0x24/0x40
   vfs_get_tree+0x22/0xb0
   vfs_kern_mount.part.0+0x71/0xb0
   btrfs_mount+0x10d/0x380
   ? vfs_parse_fs_string+0x4d/0x90
   legacy_get_tree+0x24/0x40
   vfs_get_tree+0x22/0xb0
   path_mount+0x433/0xa10
   __x64_sys_mount+0xe3/0x120
   do_syscall_64+0x3d/0x80
   entry_SYSCALL_64_after_hwframe+0x44/0xae

We can get -EIO or any number of legitimate errors from btrfs_search_slot(), panicing here is not the appropriate response. The error path for this code handles errors properly, simply return the error.

Signed-off-by Josef Bacik <[email protected]>

Reviewed-by David Sterba <[email protected]>

Signed-off-by David Sterba <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 fs/btrfs/tree-log.c | 2 --
 1 file changed, 2 deletions(-)

fs/btrfs/tree-log.c +0 -2

index de53e5166997..54647eb9c6ed 100644
@@ -1846,8 +1846,6 @@ static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans,
 		ret = btrfs_update_inode(trans, root, inode);
 	} else if (ret == -EEXIST) {
 		ret = 0;
-	} else {
-		BUG(); /* Logic Error */
 	}
 	iput(inode);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #127 lore raw

PATCH 127/177 platform/x86: hp-wireless: add AMDs hardware id to the supported list

From: Shyam Sundar S K <[email protected]>

[ Upstream commit f048630bdd55eb5379ef35f971639fe52fabe499 ]

Newer AMD based laptops uses AMDI0051 as the hardware id to support the airplane mode button. Adding this to the supported list.

Signed-off-by Shyam Sundar S K <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/platform/x86/hp-wireless.c | 2 ++
 1 file changed, 2 insertions(+)

drivers/platform/x86/hp-wireless.c +2 -0

index 12c31fd5d5ae..0753ef18e721 100644
@@ -17,12 +17,14 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Alex Hung");
 MODULE_ALIAS("acpi*:HPQ6001:*");
 MODULE_ALIAS("acpi*:WSTADEF:*");
+MODULE_ALIAS("acpi*:AMDI0051:*");
 
 static struct input_dev *hpwl_input_dev;
 
 static const struct acpi_device_id hpwl_ids[] = {
 	{"HPQ6001", 0},
 	{"WSTADEF", 0},
+	{"AMDI0051", 0},
 	{"", 0},
 };

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #128 lore raw

PATCH 128/177 platform/x86: intel_punit_ipc: Append MODULE_DEVICE_TABLE for ACPI

From: Andy Shevchenko <[email protected]>

[ Upstream commit bc1eca606d8084465e6f89fd646cc71defbad490 ]

The intel_punit_ipc driver might be compiled as a module. When udev handles the event of the devices appearing the intel_punit_ipc module is missing.

Append MODULE_DEVICE_TABLE for ACPI case to fix the loading issue.

Signed-off-by Andy Shevchenko <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/platform/x86/intel_punit_ipc.c | 1 +
 1 file changed, 1 insertion(+)

drivers/platform/x86/intel_punit_ipc.c +1 -0

index fa97834fdb78..ccb44f2eb240 100644
@@ -328,6 +328,7 @@ static const struct acpi_device_id punit_ipc_acpi_ids[] = {
 	{ "INT34D4", 0 },
 	{ }
 };
+MODULE_DEVICE_TABLE(acpi, punit_ipc_acpi_ids);
 
 static struct platform_driver intel_punit_ipc_driver = {
 	.probe = intel_punit_ipc_probe,

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #129 lore raw

PATCH 129/177 platform/x86: touchscreen_dmi: Add info for the Mediacom Winpad 7.0 W700 tablet

From: Teava Radu <[email protected]>

[ Upstream commit 39a6172ea88b3117353ae16cbb0a53cd80a9340a ]

Add touchscreen info for the Mediacom Winpad 7.0 W700 tablet. Tested on 5.11 hirsute.
Note: it's hw clone to Wintron surftab 7.

Signed-off-by Teava Radu <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/platform/x86/touchscreen_dmi.c | 8 ++++++++
 1 file changed, 8 insertions(+)

drivers/platform/x86/touchscreen_dmi.c +8 -0

index 7ed1189a7200..515c66ca1aec 100644
@@ -838,6 +838,14 @@ static const struct dmi_system_id touchscreen_dmi_table[] = {
 			DMI_MATCH(DMI_BIOS_VERSION, "jumperx.T87.KFBNEEA"),
 		},
 	},
+	{
+		/* Mediacom WinPad 7.0 W700 (same hw as Wintron surftab 7") */
+		.driver_data = (void *)&trekstor_surftab_wintron70_data,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "MEDIACOM"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "WinPad 7 W10 - WPW700"),
+		},
+	},
 	{
 		/* Mediacom Flexbook Edge 11 (same hw as TS Primebook C11) */
 		.driver_data = (void *)&trekstor_primebook_c11_data,

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #130 lore raw

PATCH 130/177 SMB3: incorrect file id in requests compounded with open

From: Steve French <[email protected]>

[ Upstream commit c0d46717b95735b0eacfddbcca9df37a49de9c7a ]

See MS-SMB2 3.2.4.1.4, file ids in compounded requests should be set to 0xFFFFFFFFFFFFFFFF (we were treating it as u32 not u64 and setting it incorrectly).

Signed-off-by Steve French <[email protected]>

Reported-by Stefan Metzmacher <[email protected]>

Reviewed-by Shyam Prasad N <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 fs/cifs/smb2pdu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

fs/cifs/smb2pdu.c +3 -3

index c02983ebd763..e068f82ffedd 100644
@@ -3611,10 +3611,10 @@ smb2_new_read_req(void **buf, unsigned int *total_len,
 			 * Related requests use info from previous read request
 			 * in chain.
 			 */
-			shdr->SessionId = 0xFFFFFFFF;
+			shdr->SessionId = 0xFFFFFFFFFFFFFFFF;
 			shdr->TreeId = 0xFFFFFFFF;
-			req->PersistentFileId = 0xFFFFFFFF;
-			req->VolatileFileId = 0xFFFFFFFF;
+			req->PersistentFileId = 0xFFFFFFFFFFFFFFFF;
+			req->VolatileFileId = 0xFFFFFFFFFFFFFFFF;
 		}
 	}
 	if (remaining_bytes > io_parms->length)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #131 lore raw

PATCH 131/177 drm/amd/display: Disconnect non-DP with no EDID

From: Chris Park <[email protected]>

[ Upstream commit 080039273b126eeb0185a61c045893a25dbc046e ]

[Why]
Active DP dongles return no EDID when dongle is connected, but VGA display is taken out. Current driver behavior does not remove the active display when this happens, and this is a gap between dongle DTP and dongle behavior.

[How]
For active DP dongles and non-DP scenario,
disconnect sink on detection when no EDID
is read due to timeout.

Signed-off-by Chris Park <[email protected]>

Reviewed-by Nicholas Kazlauskas <[email protected]>

Acked-by Stylon Wang <[email protected]>

Tested-by Daniel Wheeler <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpu/drm/amd/display/dc/core/dc_link.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

drivers/gpu/drm/amd/display/dc/core/dc_link.c +18 -0

index 40041c61a100..6b03267021ea 100644
@@ -936,6 +936,24 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
 			    dc_is_dvi_signal(link->connector_signal)) {
 				if (prev_sink != NULL)
 					dc_sink_release(prev_sink);
+				link_disconnect_sink(link);
+
+				return false;
+			}
+			/*
+			 * Abort detection for DP connectors if we have
+			 * no EDID and connector is active converter
+			 * as there are no display downstream
+			 *
+			 */
+			if (dc_is_dp_sst_signal(link->connector_signal) &&
+				(link->dpcd_caps.dongle_type ==
+						DISPLAY_DONGLE_DP_VGA_CONVERTER ||
+				link->dpcd_caps.dongle_type ==
+						DISPLAY_DONGLE_DP_DVI_CONVERTER)) {
+				if (prev_sink)
+					dc_sink_release(prev_sink);
+				link_disconnect_sink(link);
 
 				return false;
 			}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #132 lore raw

PATCH 132/177 drm/amd/amdgpu: fix refcount leak

From: Jingwen Chen <[email protected]>

[ Upstream commit fa7e6abc75f3d491bc561734312d065dc9dc2a77 ]

[Why]
the gem object rfb->base.obj[0] is get according to num_planes in amdgpufb_create, but is not put according to num_planes

[How]
put rfb->base.obj[0] in amdgpu_fbdev_destroy according to num_planes

Signed-off-by Jingwen Chen <[email protected]>

Reviewed-by Alex Deucher <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c | 3 +++
 1 file changed, 3 insertions(+)

drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c +3 -0

index fd94a17fb2c6..46522804c7d8 100644
@@ -289,10 +289,13 @@ out:
 static int amdgpu_fbdev_destroy(struct drm_device *dev, struct amdgpu_fbdev *rfbdev)
 {
 	struct amdgpu_framebuffer *rfb = &rfbdev->rfb;
+	int i;
 
 	drm_fb_helper_unregister_fbi(&rfbdev->helper);
 
 	if (rfb->base.obj[0]) {
+		for (i = 0; i < rfb->base.format->num_planes; i++)
+			drm_gem_object_put(rfb->base.obj[0]);
 		amdgpufb_destroy_pinned_object(rfb->base.obj[0]);
 		rfb->base.obj[0] = NULL;
 		drm_framebuffer_unregister_private(&rfb->base);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #133 lore raw

PATCH 133/177 drm/amdgpu: Fix a use-after-free

From: xinhui pan <[email protected]>

[ Upstream commit 1e5c37385097c35911b0f8a0c67ffd10ee1af9a2 ]

looks like we forget to set ttm->sg to NULL.
Hit panic below

[ 1235.844104] general protection fault, probably for non-canonical address 0x6b6b6b6b6b6b7b4b: 0000 [#1] SMP DEBUG_PAGEALLOC NOPTI
[ 1235.989074] Call Trace:
[ 1235.991751]  sg_free_table+0x17/0x20
[ 1235.995667]  amdgpu_ttm_backend_unbind.cold+0x4d/0xf7 [amdgpu]
[ 1236.002288]  amdgpu_ttm_backend_destroy+0x29/0x130 [amdgpu]
[ 1236.008464]  ttm_tt_destroy+0x1e/0x30 [ttm]
[ 1236.013066]  ttm_bo_cleanup_memtype_use+0x51/0xa0 [ttm]
[ 1236.018783]  ttm_bo_release+0x262/0xa50 [ttm]
[ 1236.023547]  ttm_bo_put+0x82/0xd0 [ttm]
[ 1236.027766]  amdgpu_bo_unref+0x26/0x50 [amdgpu]
[ 1236.032809]  amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu+0x7aa/0xd90 [amdgpu]
[ 1236.040400]  kfd_ioctl_alloc_memory_of_gpu+0xe2/0x330 [amdgpu]
[ 1236.046912]  kfd_ioctl+0x463/0x690 [amdgpu]

Signed-off-by xinhui pan <[email protected]>

Reviewed-by Christian König <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 1 +
 1 file changed, 1 insertion(+)

drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +1 -0

index 91e3a87b1de8..58e14d3040f0 100644
@@ -1300,6 +1300,7 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_tt *ttm)
 	if (gtt && gtt->userptr) {
 		amdgpu_ttm_tt_set_user_pages(ttm, NULL);
 		kfree(ttm->sg);
+		ttm->sg = NULL;
 		ttm->page_flags &= ~TTM_PAGE_FLAG_SG;
 		return;
 	}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #134 lore raw

PATCH 134/177 drm/amd/amdgpu: fix a potential deadlock in gpu reset

From: Lang Yu <[email protected]>

[ Upstream commit 9c2876d56f1ce9b6b2072f1446fb1e8d1532cb3d ]

When amdgpu_ib_ring_tests failed, the reset logic called amdgpu_device_ip_suspend twice, then deadlock occurred. Deadlock log:

[  805.655192] amdgpu 0000:04:00.0: amdgpu: ib ring test failed (-110).
[  806.290952] [drm] free PSP TMR buffer

[  806.319406] ============================================
[  806.320315] WARNING: possible recursive locking detected
[  806.321225] 5.11.0-custom #1 Tainted: G        W  OEL
[  806.322135] --------------------------------------------
[  806.323043] cat/2593 is trying to acquire lock:
[  806.323825] ffff888136b1cdc8 (&adev->dm.dc_lock){+.+.}-{3:3}, at: dm_suspend+0xb8/0x1d0 [amdgpu]
[  806.325668]
               but task is already holding lock:
[  806.326664] ffff888136b1cdc8 (&adev->dm.dc_lock){+.+.}-{3:3}, at: dm_suspend+0xb8/0x1d0 [amdgpu]
[  806.328430]
               other info that might help us debug this:
[  806.329539]  Possible unsafe locking scenario:

[  806.330549]        CPU0
[  806.330983]        ----
[  806.331416]   lock(&adev->dm.dc_lock);
[  806.332086]   lock(&adev->dm.dc_lock);
[  806.332738]
                *** DEADLOCK ***

[  806.333747]  May be due to missing lock nesting notation

[  806.334899] 3 locks held by cat/2593:
[  806.335537]  #0: ffff888100d3f1b8 (&attr->mutex){+.+.}-{3:3}, at: simple_attr_read+0x4e/0x110
[  806.337009]  #1: ffff888136b1fd78 (&adev->reset_sem){++++}-{3:3}, at: amdgpu_device_lock_adev+0x42/0x94 [amdgpu]
[  806.339018]  #2: ffff888136b1cdc8 (&adev->dm.dc_lock){+.+.}-{3:3}, at: dm_suspend+0xb8/0x1d0 [amdgpu]
[  806.340869]
               stack backtrace:
[  806.341621] CPU: 6 PID: 2593 Comm: cat Tainted: G        W  OEL    5.11.0-custom #1
[  806.342921] Hardware name: AMD Celadon-CZN/Celadon-CZN, BIOS WLD0C23N_Weekly_20_12_2 12/23/2020
[  806.344413] Call Trace:
[  806.344849]  dump_stack+0x93/0xbd
[  806.345435]  __lock_acquire.cold+0x18a/0x2cf
[  806.346179]  lock_acquire+0xca/0x390
[  806.346807]  ? dm_suspend+0xb8/0x1d0 [amdgpu]
[  806.347813]  __mutex_lock+0x9b/0x930
[  806.348454]  ? dm_suspend+0xb8/0x1d0 [amdgpu]
[  806.349434]  ? amdgpu_device_indirect_rreg+0x58/0x70 [amdgpu]
[  806.350581]  ? _raw_spin_unlock_irqrestore+0x47/0x50
[  806.351437]  ? dm_suspend+0xb8/0x1d0 [amdgpu]
[  806.352437]  ? rcu_read_lock_sched_held+0x4f/0x80
[  806.353252]  ? rcu_read_lock_sched_held+0x4f/0x80
[  806.354064]  mutex_lock_nested+0x1b/0x20
[  806.354747]  ? mutex_lock_nested+0x1b/0x20
[  806.355457]  dm_suspend+0xb8/0x1d0 [amdgpu]
[  806.356427]  ? soc15_common_set_clockgating_state+0x17d/0x19 [amdgpu]
[  806.357736]  amdgpu_device_ip_suspend_phase1+0x78/0xd0 [amdgpu]
[  806.360394]  amdgpu_device_ip_suspend+0x21/0x70 [amdgpu]
[  806.362926]  amdgpu_device_pre_asic_reset+0xb3/0x270 [amdgpu]
[  806.365560]  amdgpu_device_gpu_recover.cold+0x679/0x8eb [amdgpu]

Signed-off-by Lang Yu <[email protected]>

Acked-by Christian KÃnig <[email protected]>

Reviewed-by Andrey Grodzovsky <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 1 -
 1 file changed, 1 deletion(-)

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +0 -1

index 3b3fc9a426e9..765f9a6c4640 100644
@@ -3704,7 +3704,6 @@ out:
 			r = amdgpu_ib_ring_tests(tmp_adev);
 			if (r) {
 				dev_err(tmp_adev->dev, "ib ring test failed (%d).\n", r);
-				r = amdgpu_device_ip_suspend(tmp_adev);
 				need_full_reset = true;
 				r = -EAGAIN;
 				goto end;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #135 lore raw

PATCH 135/177 net: netcp: Fix an error message

From: Christophe JAILLET <[email protected]>

[ Upstream commit ddb6e00f8413e885ff826e32521cff7924661de0 ]

'ret' is known to be 0 here.
The expected error code is stored in 'tx_pipe->dma_queue', so use it instead.

While at it, switch from %d to %pe which is more user friendly.

Fixes 84640e27f230 ("net: netcp: Add Keystone NetCP core ethernet driver")

Signed-off-by Christophe JAILLET <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/ti/netcp_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/net/ethernet/ti/netcp_core.c +2 -2

index 1b2702f74455..c0025bb8a584 100644
@@ -1350,8 +1350,8 @@ int netcp_txpipe_open(struct netcp_tx_pipe *tx_pipe)
 	tx_pipe->dma_queue = knav_queue_open(name, tx_pipe->dma_queue_id,
 					     KNAV_QUEUE_SHARED);
 	if (IS_ERR(tx_pipe->dma_queue)) {
-		dev_err(dev, "Could not open DMA queue for channel \"%s\": %d\n",
-			name, ret);
+		dev_err(dev, "Could not open DMA queue for channel \"%s\": %pe\n",
+			name, tx_pipe->dma_queue);
 		ret = PTR_ERR(tx_pipe->dma_queue);
 		goto err;
 	}

Signature · 1 line

2.30.2

Marion & Christophe JAILLET

2021-05-31 18:46 UTC #136 lore raw

Re: [PATCH 5.4 135/177] net: netcp: Fix an error message

Le 31/05/2021 à 15:14, Greg Kroah-Hartman a écrit :

Quoted 23 lines

From: Christophe JAILLET <[email protected]>
[ Upstream commit ddb6e00f8413e885ff826e32521cff7924661de0 ]
'ret' is known to be 0 here.
The expected error code is stored in 'tx_pipe->dma_queue', so use it instead.
While at it, switch from %d to %pe which is more user friendly.
Fixes 84640e27f230 ("net: netcp: Add Keystone NetCP core ethernet driver")
Signed-off-by Christophe JAILLET <[email protected]>
Signed-off-by David S. Miller <[email protected]>
Signed-off-by Sasha Levin <[email protected]>
---
  drivers/net/ethernet/ti/netcp_core.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/ti/netcp_core.c +0 -0
index 1b2702f74455..c0025bb8a584 100644

drivers/net/ethernet/ti/netcp_core.c int netcp_txpipe_open(struct netcp_tx_pipe *tx_pipe) · line 1357

Quoted 10 lines

drivers/net/ethernet/ti/netcp_core.c +2 -2

@@ -1350,8 +1350,8 @@ int netcp_txpipe_open(struct netcp_tx_pipe *tx_pipe)
  	tx_pipe->dma_queue = knav_queue_open(name, tx_pipe->dma_queue_id,
  					     KNAV_QUEUE_SHARED);
  	if (IS_ERR(tx_pipe->dma_queue)) {
-		dev_err(dev, "Could not open DMA queue for channel \"%s\": %d\n",
-			name, ret);
+		dev_err(dev, "Could not open DMA queue for channel \"%s\": %pe\n",
+			name, tx_pipe->dma_queue);
  		ret = PTR_ERR(tx_pipe->dma_queue);
  		goto err;
  	}

Hi,

Apparently %pe is only supported up to (including) 5.5. It is not part of 5.4.123.

So this patch should not be backported here or should be backported differently, ie:
leave dev_err as-is
move "ret = PTR_ERR(tx_pipe->dma_queue);" 1 line above

(or %pe should be backported first)

PS: adding Dan Carpenter because we had a small discussion about some potential backport issue when, using %pe

Sasha Levin

maintainer

2021-05-31 21:42 UTC #137 lore raw

Re: [PATCH 5.4 135/177] net: netcp: Fix an error message

Quoted 39 lines · Marion & Christophe JAILLET

Le 31/05/2021 à 15:14, Greg Kroah-Hartman a écrit :
Quoted 23 lines
From: Christophe JAILLET <[email protected]>
[ Upstream commit ddb6e00f8413e885ff826e32521cff7924661de0 ]
'ret' is known to be 0 here.
The expected error code is stored in 'tx_pipe->dma_queue', so use it instead.
While at it, switch from %d to %pe which is more user friendly.
Fixes 84640e27f230 ("net: netcp: Add Keystone NetCP core ethernet driver")
Signed-off-by Christophe JAILLET <[email protected]>
Signed-off-by David S. Miller <[email protected]>
Signed-off-by Sasha Levin <[email protected]>
---
 drivers/net/ethernet/ti/netcp_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/ti/netcp_core.c +0 -0
index 1b2702f74455..c0025bb8a584 100644
drivers/net/ethernet/ti/netcp_core.c int netcp_txpipe_open(struct netcp_tx_pipe *tx_pipe) · line 1357
Quoted 10 lines
drivers/net/ethernet/ti/netcp_core.c +2 -2
@@ -1350,8 +1350,8 @@ int netcp_txpipe_open(struct netcp_tx_pipe *tx_pipe)
 	tx_pipe->dma_queue = knav_queue_open(name, tx_pipe->dma_queue_id,
 					     KNAV_QUEUE_SHARED);
 	if (IS_ERR(tx_pipe->dma_queue)) {
-		dev_err(dev, "Could not open DMA queue for channel \"%s\": %d\n",
-			name, ret);
+		dev_err(dev, "Could not open DMA queue for channel \"%s\": %pe\n",
+			name, tx_pipe->dma_queue);
 		ret = PTR_ERR(tx_pipe->dma_queue);
 		goto err;
 	}
Hi,

Quoted 5 lines

Apparently %pe is only supported up to (including) 5.5. It is not part of 5.4.123.
So this patch should not be backported here or should be backported differently, ie:
leave dev_err as-is
move "ret = PTR_ERR(tx_pipe->dma_queue);" 1 line above

I'll fix it up the way you've described above. Thanks!

Signature · 2 lines

Thanks,
Sasha

Dan Carpenter

2021-06-01 07:09 UTC #138 lore raw

Re: [PATCH 5.4 135/177] net: netcp: Fix an error message

Quoted 5 lines · Sasha Levin

Quoted 5 lines
Hi,
Apparently %pe is only supported up to (including) 5.5. It is not part of 5.4.123.

It seems like it would be easier to just back port the %pe commit to v5.4? It seems to apply and build basically fine? I didn't try test it. It's commit 57f5677e535b ("printf: add support for printing symbolic error names").

Rasmus, we were always wondering when you were going to make it work for integer error codes. ;)

regards,
dan carpenter

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #139 lore raw

PATCH 136/177 net: dsa: fix error code getting shifted with 4 in dsa_slave_get_sset_count

From: Vladimir Oltean <[email protected]>

[ Upstream commit b94cbc909f1d80378a1f541968309e5c1178c98b ]

DSA implements a bunch of 'standardized' ethtool statistics counters, namely tx_packets, tx_bytes, rx_packets, rx_bytes. So whatever the hardware driver returns in .get_sset_count(), we need to add 4 to that.

That is ok, except that .get_sset_count() can return a negative error code, for example:

b53_get_sset_count
-> phy_ethtool_get_sset_count
-> return -EIO

-EIO is -5, and with 4 added to it, it becomes -1, aka -EPERM. One can imagine that certain error codes may even become positive, although based on code inspection I did not see instances of that.

Check the error code first, if it is negative return it as-is.

Based on a similar patch for dsa_master_get_strings from Dan Carpenter:
https://patchwork.kernel.org/project/netdevbpf/patch/YJaSe3RPgn7gKxZv@mwanda/

Fixes 91da11f870f0 ("net: Distributed Switch Architecture protocol support")

Signed-off-by Vladimir Oltean <[email protected]>

Reviewed-by Florian Fainelli <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/dsa/slave.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

net/dsa/slave.c +7 -5

index 06f8874d53ee..75b4cd4bcafb 100644
@@ -692,13 +692,15 @@ static int dsa_slave_get_sset_count(struct net_device *dev, int sset)
 	struct dsa_switch *ds = dp->ds;
 
 	if (sset == ETH_SS_STATS) {
-		int count;
+		int count = 0;
 
-		count = 4;
-		if (ds->ops->get_sset_count)
-			count += ds->ops->get_sset_count(ds, dp->index, sset);
+		if (ds->ops->get_sset_count) {
+			count = ds->ops->get_sset_count(ds, dp->index, sset);
+			if (count < 0)
+				return count;
+		}
 
-		return count;
+		return count + 4;
 	}
 
 	return -EOPNOTSUPP;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:26 UTC #140 lore raw

PATCH 137/177 ASoC: cs42l42: Regmap must use_single_read/write

From: Richard Fitzgerald <[email protected]>

[ Upstream commit 0fad605fb0bdc00d8ad78696300ff2fbdee6e048 ]

cs42l42 does not support standard burst transfers so the use_single_read and use_single_write flags must be set in the regmap config.

Because of this bug, the patch:

commit 0a0eb567e1d4 ("ASoC: cs42l42: Minor error paths fixups")

broke cs42l42 probe() because without the use_single_* flags it causes regmap to issue a burst read.

However, the missing use_single_* could cause problems anyway because the regmap cache can attempt burst transfers if these flags are not set.

Fixes 2c394ca79604 ("ASoC: Add support for CS42L42 codec")

Signed-off-by Richard Fitzgerald <[email protected]>

Acked-by Charles Keepax <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Mark Brown <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 sound/soc/codecs/cs42l42.c | 3 +++
 1 file changed, 3 insertions(+)

sound/soc/codecs/cs42l42.c +3 -0

index dcd2acb2c3ce..5faf8877137a 100644
@@ -398,6 +398,9 @@ static const struct regmap_config cs42l42_regmap = {
 	.reg_defaults = cs42l42_reg_defaults,
 	.num_reg_defaults = ARRAY_SIZE(cs42l42_reg_defaults),
 	.cache_type = REGCACHE_RBTREE,
+
+	.use_single_read = true,
+	.use_single_write = true,
 };
 
 static DECLARE_TLV_DB_SCALE(adc_tlv, -9600, 100, false);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:25 UTC #141 lore raw

PATCH 138/177 vfio-ccw: Check initialized flag in cp_init()

From: Eric Farman <[email protected]>

[ Upstream commit c6c82e0cd8125d30f2f1b29205c7e1a2f1a6785b ]

We have a really nice flag in the channel_program struct that indicates if it had been initialized by cp_init(), and use it as a guard in the other cp accessor routines, but not for a duplicate call into cp_init(). The possibility of this occurring is low, because that flow is protected by the private->io_mutex and FSM CP_PROCESSING state. But then why bother checking it in (for example) cp_prefetch() then?

Let's just be consistent and check for that in cp_init() too.

Fixes 71189f263f8a3 ("vfio-ccw: make it safe to access channel programs")

Signed-off-by Eric Farman <[email protected]>

Reviewed-by Cornelia Huck <[email protected]>

Acked-by Matthew Rosato <[email protected]>

Message-Id: <[email protected]>

Signed-off-by Cornelia Huck <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/s390/cio/vfio_ccw_cp.c | 4 ++++
 1 file changed, 4 insertions(+)

drivers/s390/cio/vfio_ccw_cp.c +4 -0

index 3645d1720c4b..9628e0f3add3 100644
@@ -636,6 +636,10 @@ int cp_init(struct channel_program *cp, struct device *mdev, union orb *orb)
 {
 	int ret;
 
+	/* this is an error in the caller */
+	if (cp->initialized)
+		return -EBUSY;
+
 	/*
 	 * XXX:
 	 * Only support prefetch enable mode now.

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:26 UTC #142 lore raw

PATCH 139/177 net: really orphan skbs tied to closing sk

From: Paolo Abeni <[email protected]>

[ Upstream commit 098116e7e640ba677d9e345cbee83d253c13d556 ]

If the owing socket is shutting down - e.g. the sock reference count already dropped to 0 and only sk_wmem_alloc is keeping the sock alive, skb_orphan_partial() becomes a no-op.

When forwarding packets over veth with GRO enabled, the above causes refcount errors.

This change addresses the issue with a plain skb_orphan() call in the critical scenario.

Fixes 9adc89af724f ("net: let skb_orphan_partial wake-up waiters.")

Signed-off-by Paolo Abeni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/sock.h | 4 +++-
 net/core/sock.c    | 8 ++++----
 2 files changed, 7 insertions(+), 5 deletions(-)

include/net/sock.h +3 -1

index 4137fa178790..a0728f24ecc5 100644
@@ -2150,13 +2150,15 @@ static inline void skb_set_owner_r(struct sk_buff *skb, struct sock *sk)
 	sk_mem_charge(sk, skb->truesize);
 }
 
-static inline void skb_set_owner_sk_safe(struct sk_buff *skb, struct sock *sk)
+static inline __must_check bool skb_set_owner_sk_safe(struct sk_buff *skb, struct sock *sk)
 {
 	if (sk && refcount_inc_not_zero(&sk->sk_refcnt)) {
 		skb_orphan(skb);
 		skb->destructor = sock_efree;
 		skb->sk = sk;
+		return true;
 	}
+	return false;
 }
 
 void sk_reset_timer(struct sock *sk, struct timer_list *timer,

net/core/sock.c +4 -4

index 19c178aac0ae..68f84fac63e0 100644
@@ -2026,10 +2026,10 @@ void skb_orphan_partial(struct sk_buff *skb)
 	if (skb_is_tcp_pure_ack(skb))
 		return;
 
-	if (can_skb_orphan_partial(skb))
-		skb_set_owner_sk_safe(skb, skb->sk);
-	else
-		skb_orphan(skb);
+	if (can_skb_orphan_partial(skb) && skb_set_owner_sk_safe(skb, skb->sk))
+		return;
+
+	skb_orphan(skb);
 }
 EXPORT_SYMBOL(skb_orphan_partial);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:26 UTC #143 lore raw

PATCH 140/177 net: fec: fix the potential memory leak in fec_enet_init()

From: Fugang Duan <[email protected]>

[ Upstream commit 619fee9eb13b5d29e4267cb394645608088c28a8 ]

If the memory allocated for cbd_base is failed, it should free the memory allocated for the queues, otherwise it causes memory leak.

And if the memory allocated for the queues is failed, it can return error directly.

Fixes 59d0f7465644 ("net: fec: init multi queue date structure")

Signed-off-by Fugang Duan <[email protected]>

Signed-off-by Joakim Zhang <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/freescale/fec_main.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

drivers/net/ethernet/freescale/fec_main.c +9 -2

index fd7fc6f20c9d..b1856552ab81 100644
@@ -3274,7 +3274,9 @@ static int fec_enet_init(struct net_device *ndev)
 		return ret;
 	}
 
-	fec_enet_alloc_queue(ndev);
+	ret = fec_enet_alloc_queue(ndev);
+	if (ret)
+		return ret;
 
 	bd_size = (fep->total_tx_ring_size + fep->total_rx_ring_size) * dsize;

@@ -3282,7 +3284,8 @@ static int fec_enet_init(struct net_device *ndev)
 	cbd_base = dmam_alloc_coherent(&fep->pdev->dev, bd_size, &bd_dma,
 				       GFP_KERNEL);
 	if (!cbd_base) {
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto free_queue_mem;
 	}
 
 	/* Get the Ethernet address */

@@ -3360,6 +3363,10 @@ static int fec_enet_init(struct net_device *ndev)
 		fec_enet_update_ethtool_stats(ndev);
 
 	return 0;
+
+free_queue_mem:
+	fec_enet_free_queue(ndev);
+	return ret;
 }
 
 #ifdef CONFIG_OF

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:26 UTC #144 lore raw

PATCH 141/177 net: mdio: thunder: Fix a double free issue in the .remove function

From: Christophe JAILLET <[email protected]>

[ Upstream commit a93a0a15876d2a077a3bc260b387d2457a051f24 ]

'bus->mii_bus' have been allocated with 'devm_mdiobus_alloc_size()' in the probe function. So it must not be freed explicitly or there will be a double free.

Remove the incorrect 'mdiobus_free' in the remove function.

Fixes 379d7ac7ca31 ("phy: mdio-thunder: Add driver for Cavium Thunder SoC MDIO buses.")

Signed-off-by Christophe JAILLET <[email protected]>

Reviewed-by Russell King <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/phy/mdio-thunder.c | 1 -
 1 file changed, 1 deletion(-)

drivers/net/phy/mdio-thunder.c +0 -1

index b6128ae7f14f..1e2f57ed1ef7 100644
@@ -126,7 +126,6 @@ static void thunder_mdiobus_pci_remove(struct pci_dev *pdev)
 			continue;
 
 		mdiobus_unregister(bus->mii_bus);
-		mdiobus_free(bus->mii_bus);
 		oct_mdio_writeq(0, bus->register_base + SMI_EN);
 	}
 	pci_set_drvdata(pdev, NULL);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #145 lore raw

PATCH 142/177 net: mdio: octeon: Fix some double free issues

From: Christophe JAILLET <[email protected]>

[ Upstream commit e1d027dd97e1e750669cdc0d3b016a4f54e473eb ]

'bus->mii_bus' has been allocated with 'devm_mdiobus_alloc_size()' in the probe function. So it must not be freed explicitly or there will be a double free.

Remove the incorrect 'mdiobus_free' in the error handling path of the probe function and in remove function.

Fixes 35d2aeac9810 ("phy: mdio-octeon: Use devm_mdiobus_alloc_size()")

Signed-off-by Christophe JAILLET <[email protected]>

Reviewed-by Russell King <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/phy/mdio-octeon.c | 2 --
 1 file changed, 2 deletions(-)

drivers/net/phy/mdio-octeon.c +0 -2

index 8327382aa568..088c73731652 100644
@@ -72,7 +72,6 @@ static int octeon_mdiobus_probe(struct platform_device *pdev)
 
 	return 0;
 fail_register:
-	mdiobus_free(bus->mii_bus);
 	smi_en.u64 = 0;
 	oct_mdio_writeq(smi_en.u64, bus->register_base + SMI_EN);
 	return err;

@@ -86,7 +85,6 @@ static int octeon_mdiobus_remove(struct platform_device *pdev)
 	bus = platform_get_drvdata(pdev);
 
 	mdiobus_unregister(bus->mii_bus);
-	mdiobus_free(bus->mii_bus);
 	smi_en.u64 = 0;
 	oct_mdio_writeq(smi_en.u64, bus->register_base + SMI_EN);
 	return 0;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #146 lore raw

PATCH 143/177 openvswitch: meter: fix race when getting now_ms.

[1] https://patchwork.ozlabs.org/project/openvswitch/patch/[email protected]/

From: Tao Liu <[email protected]>

[ Upstream commit e4df1b0c24350a0f00229ff895a91f1072bd850d ]

We have observed meters working unexpected if traffic is 3+Gbit/s with multiple connections.

now_ms is not pretected by meter->lock, we may get a negative long_delta_ms when another cpu updated meter->used, then:

    delta_ms = (u32)long_delta_ms;

which will be a large value.

    band->bucket += delta_ms * band->rate;

then we get a wrong band->bucket.

OpenVswitch userspace datapath has fixed the same issue[1] some time ago, and we port the implementation to kernel datapath.

Fixes 96fbc13d7e77 ("openvswitch: Add meter infrastructure")

Signed-off-by Tao Liu <[email protected]>

Reviewed-by Ilya Maximets <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/openvswitch/meter.c | 8 ++++++++
 1 file changed, 8 insertions(+)

net/openvswitch/meter.c +8 -0

index 541eea74ef7a..c37e09223cbb 100644
@@ -460,6 +460,14 @@ bool ovs_meter_execute(struct datapath *dp, struct sk_buff *skb,
 	spin_lock(&meter->lock);
 
 	long_delta_ms = (now_ms - meter->used); /* ms */
+	if (long_delta_ms < 0) {
+		/* This condition means that we have several threads fighting
+		 * for a meter lock, and the one who received the packets a
+		 * bit later wins. Assuming that all racing threads received
+		 * packets at the same time to avoid overflow.
+		 */
+		long_delta_ms = 0;
+	}
 
 	/* Make sure delta_ms will not be too large, so that bucket will not
 	 * wrap around below.

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #147 lore raw

PATCH 144/177 tls splice: check SPLICE_F_NONBLOCK instead of MSG_DONTWAIT

From: Jim Ma <[email protected]>

[ Upstream commit 974271e5ed45cfe4daddbeb16224a2156918530e ]

In tls_sw_splice_read, checkout MSG_* is inappropriate, should use SPLICE_*, update tls_wait_data to accept nonblock arguments instead of flags for recvmsg and splice.

Fixes c46234ebb4d1 ("tls: RX path for ktls")

Signed-off-by Jim Ma <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/tls/tls_sw.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

net/tls/tls_sw.c +6 -5

index 0d524ef0d8c8..cdb65aa54be7 100644
@@ -37,6 +37,7 @@
 
 #include <linux/sched/signal.h>
 #include <linux/module.h>
+#include <linux/splice.h>
 #include <crypto/aead.h>
 
 #include <net/strparser.h>

@@ -1278,7 +1279,7 @@ int tls_sw_sendpage(struct sock *sk, struct page *page,
 }
 
 static struct sk_buff *tls_wait_data(struct sock *sk, struct sk_psock *psock,
-				     int flags, long timeo, int *err)
+				     bool nonblock, long timeo, int *err)
 {
 	struct tls_context *tls_ctx = tls_get_ctx(sk);
 	struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);

@@ -1303,7 +1304,7 @@ static struct sk_buff *tls_wait_data(struct sock *sk, struct sk_psock *psock,
 		if (sock_flag(sk, SOCK_DONE))
 			return NULL;
 
-		if ((flags & MSG_DONTWAIT) || !timeo) {
+		if (nonblock || !timeo) {
 			*err = -EAGAIN;
 			return NULL;
 		}

@@ -1781,7 +1782,7 @@ int tls_sw_recvmsg(struct sock *sk,
 		bool async_capable;
 		bool async = false;
 
-		skb = tls_wait_data(sk, psock, flags, timeo, &err);
+		skb = tls_wait_data(sk, psock, flags & MSG_DONTWAIT, timeo, &err);
 		if (!skb) {
 			if (psock) {
 				int ret = __tcp_bpf_recvmsg(sk, psock,

@@ -1985,9 +1986,9 @@ ssize_t tls_sw_splice_read(struct socket *sock,  loff_t *ppos,
 
 	lock_sock(sk);
 
-	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+	timeo = sock_rcvtimeo(sk, flags & SPLICE_F_NONBLOCK);
 
-	skb = tls_wait_data(sk, NULL, flags, timeo, &err);
+	skb = tls_wait_data(sk, NULL, flags & SPLICE_F_NONBLOCK, timeo, &err);
 	if (!skb)
 		goto splice_read_end;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #148 lore raw

PATCH 145/177 net: sched: fix packet stuck problem for lockless qdisc

From: Yunsheng Lin <[email protected]>

[ Upstream commit a90c57f2cedd52a511f739fb55e6244e22e1a2fb ]

Lockless qdisc has below concurrent problem:
cpu0 cpu1
. .

q->enqueue                 .
     .                     .
qdisc_run_begin()          .
     .                     .
dequeue_skb()              .
     .                     .
sch_direct_xmit()          .
     .                     .

. q->enqueue
. qdisc_run_begin()
. return and do nothing
. .
qdisc_run_end() .

cpu1 enqueue a skb without calling __qdisc_run() because cpu0 has not released the lock yet and spin_trylock() return false for cpu1 in qdisc_run_begin(), and cpu0 do not see the skb enqueued by cpu1 when calling dequeue_skb() because cpu1 may enqueue the skb after cpu0 calling dequeue_skb() and before cpu0 calling qdisc_run_end().

Lockless qdisc has below another concurrent problem when tx_action is involved:

cpu0(serving tx_action) cpu1 cpu2

          .                   .                .
          .              q->enqueue            .
          .            qdisc_run_begin()       .
          .              dequeue_skb()         .
          .                   .            q->enqueue
          .                   .                .
          .             sch_direct_xmit()      .
          .                   .         qdisc_run_begin()
          .                   .       return and do nothing
          .                   .                .
 clear __QDISC_STATE_SCHED    .                .
 qdisc_run_begin()            .                .
 return and do nothing        .                .
          .                   .                .
          .            qdisc_run_end()         .

This patch fixes the above data race by:
1. If the first spin_trylock() return false and STATE_MISSED is not set, set STATE_MISSED and retry another spin_trylock() in case other CPU may not see STATE_MISSED after it releases the lock.
2. reschedule if STATE_MISSED is set after the lock is released at the end of qdisc_run_end().

For tx_action case, STATE_MISSED is also set when cpu1 is at the end if qdisc_run_end(), so tx_action will be rescheduled again to dequeue the skb enqueued by cpu2.

Clear STATE_MISSED before retrying a dequeuing when dequeuing returns NULL in order to reduce the overhead of the second spin_trylock() and __netif_schedule() calling.

Also clear the STATE_MISSED before calling __netif_schedule() at the end of qdisc_run_end() to avoid doing another round of dequeuing in the pfifo_fast_dequeue().

The performance impact of this patch, tested using pktgen and dummy netdev with pfifo_fast qdisc attached:

 threads  without+this_patch   with+this_patch      delta
    1        2.61Mpps            2.60Mpps           -0.3%
    2        3.97Mpps            3.82Mpps           -3.7%
    4        5.62Mpps            5.59Mpps           -0.5%
    8        2.78Mpps            2.77Mpps           -0.3%
   16        2.22Mpps            2.22Mpps           -0.0%

Fixes 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking")

Acked-by Jakub Kicinski <[email protected]>

Tested-by Juergen Gross <[email protected]>

Signed-off-by Yunsheng Lin <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/sch_generic.h | 35 ++++++++++++++++++++++++++++++++++-
 net/sched/sch_generic.c   | 19 +++++++++++++++++++
 2 files changed, 53 insertions(+), 1 deletion(-)

include/net/sch_generic.h +34 -1

index b2ceec7b280d..0852f3e51360 100644
@@ -36,6 +36,7 @@ struct qdisc_rate_table {
 enum qdisc_state_t {
 	__QDISC_STATE_SCHED,
 	__QDISC_STATE_DEACTIVATED,
+	__QDISC_STATE_MISSED,
 };
 
 struct qdisc_size_table {

@@ -156,8 +157,33 @@ static inline bool qdisc_is_empty(const struct Qdisc *qdisc)
 static inline bool qdisc_run_begin(struct Qdisc *qdisc)
 {
 	if (qdisc->flags & TCQ_F_NOLOCK) {
+		if (spin_trylock(&qdisc->seqlock))
+			goto nolock_empty;
+
+		/* If the MISSED flag is set, it means other thread has
+		 * set the MISSED flag before second spin_trylock(), so
+		 * we can return false here to avoid multi cpus doing
+		 * the set_bit() and second spin_trylock() concurrently.
+		 */
+		if (test_bit(__QDISC_STATE_MISSED, &qdisc->state))
+			return false;
+
+		/* Set the MISSED flag before the second spin_trylock(),
+		 * if the second spin_trylock() return false, it means
+		 * other cpu holding the lock will do dequeuing for us
+		 * or it will see the MISSED flag set after releasing
+		 * lock and reschedule the net_tx_action() to do the
+		 * dequeuing.
+		 */
+		set_bit(__QDISC_STATE_MISSED, &qdisc->state);
+
+		/* Retry again in case other CPU may not see the new flag
+		 * after it releases the lock at the end of qdisc_run_end().
+		 */
 		if (!spin_trylock(&qdisc->seqlock))
 			return false;
+
+nolock_empty:
 		WRITE_ONCE(qdisc->empty, false);
 	} else if (qdisc_is_running(qdisc)) {
 		return false;

@@ -173,8 +199,15 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc)
 static inline void qdisc_run_end(struct Qdisc *qdisc)
 {
 	write_seqcount_end(&qdisc->running);
-	if (qdisc->flags & TCQ_F_NOLOCK)
+	if (qdisc->flags & TCQ_F_NOLOCK) {
 		spin_unlock(&qdisc->seqlock);
+
+		if (unlikely(test_bit(__QDISC_STATE_MISSED,
+				      &qdisc->state))) {
+			clear_bit(__QDISC_STATE_MISSED, &qdisc->state);
+			__netif_schedule(qdisc);
+		}
+	}
 }
 
 static inline bool qdisc_may_bulk(const struct Qdisc *qdisc)

net/sched/sch_generic.c +19 -0

index 6e6147a81bc3..0723e7858658 100644
@@ -645,8 +645,10 @@ static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc)
 {
 	struct pfifo_fast_priv *priv = qdisc_priv(qdisc);
 	struct sk_buff *skb = NULL;
+	bool need_retry = true;
 	int band;
 
+retry:
 	for (band = 0; band < PFIFO_FAST_BANDS && !skb; band++) {
 		struct skb_array *q = band2list(priv, band);

@@ -657,6 +659,23 @@ static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc)
 	}
 	if (likely(skb)) {
 		qdisc_update_stats_at_dequeue(qdisc, skb);
+	} else if (need_retry &&
+		   test_bit(__QDISC_STATE_MISSED, &qdisc->state)) {
+		/* Delay clearing the STATE_MISSED here to reduce
+		 * the overhead of the second spin_trylock() in
+		 * qdisc_run_begin() and __netif_schedule() calling
+		 * in qdisc_run_end().
+		 */
+		clear_bit(__QDISC_STATE_MISSED, &qdisc->state);
+
+		/* Make sure dequeuing happens after clearing
+		 * STATE_MISSED.
+		 */
+		smp_mb__after_atomic();
+
+		need_retry = false;
+
+		goto retry;
 	} else {
 		WRITE_ONCE(qdisc->empty, true);
 	}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #149 lore raw

PATCH 146/177 net: sched: fix tx action rescheduling issue during deactivation

From: Yunsheng Lin <[email protected]>

[ Upstream commit 102b55ee92f9fda4dde7a45d2b20538e6e3e3d1e ]

Currently qdisc_run() checks the STATE_DEACTIVATED of lockless qdisc before calling __qdisc_run(), which ultimately clear the STATE_MISSED when all the skb is dequeued. If STATE_DEACTIVATED is set before clearing STATE_MISSED, there may be rescheduling of net_tx_action() at the end of qdisc_run_end(), see below:

CPU0(net_tx_atcion) CPU1(__dev_xmit_skb) CPU2(dev_deactivate)

          .                   .                     .
          .            set STATE_MISSED             .
          .           __netif_schedule()            .
          .                   .           set STATE_DEACTIVATED
          .                   .                qdisc_reset()
          .                   .                     .
          .<---------------   .              synchronize_net()
clear __QDISC_STATE_SCHED  |  .                     .
          .                |  .                     .
          .                |  .            some_qdisc_is_busy()
          .                |  .               return *false*
          .                |  .                     .
  test STATE_DEACTIVATED   |  .                     .
__qdisc_run() *not* called |  .                     .
          .                |  .                     .
   test STATE_MISS         |  .                     .
 __netif_schedule()--------|  .                     .
          .                   .                     .
          .                   .                     .

__qdisc_run() is not called by net_tx_atcion() in CPU0 because CPU2 has set STATE_DEACTIVATED flag during dev_deactivate(), and STATE_MISSED is only cleared in __qdisc_run(), __netif_schedule is called at the end of qdisc_run_end(), causing tx action rescheduling problem.

qdisc_run() called by net_tx_action() runs in the softirq context, which should has the same semantic as the qdisc_run() called by __dev_xmit_skb() protected by rcu_read_lock_bh(). And there is a synchronize_net() between STATE_DEACTIVATED flag being set and qdisc_reset()/some_qdisc_is_busy in dev_deactivate(), we can safely bail out for the deactived lockless qdisc in net_tx_action(), and qdisc_reset() will reset all skb not dequeued yet.

So add the rcu_read_lock() explicitly to protect the qdisc_run() and do the STATE_DEACTIVATED checking in net_tx_action() before calling qdisc_run_begin(). Another option is to do the checking in the qdisc_run_end(), but it will add unnecessary overhead for non-tx_action case, because __dev_queue_xmit() will not see qdisc with STATE_DEACTIVATED after synchronize_net(), the qdisc with STATE_DEACTIVATED can only be seen by net_tx_action() because of __netif_schedule().

The STATE_DEACTIVATED checking in qdisc_run() is to avoid race between net_tx_action() and qdisc_reset(), see:
commit d518d2ed8640 ("net/sched: fix race between deactivation and dequeue for NOLOCK qdisc"). As the bailout added above for deactived lockless qdisc in net_tx_action() provides better protection for the race without calling qdisc_run() at all, so remove the STATE_DEACTIVATED checking in qdisc_run().

After qdisc_reset(), there is no skb in qdisc to be dequeued, so clear the STATE_MISSED in dev_reset_queue() too.

Fixes 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking")

Acked-by Jakub Kicinski <[email protected]>

Signed-off-by Yunsheng Lin <[email protected]>

V8: Clearing STATE_MISSED before calling __netif_schedule() has
avoid the endless rescheduling problem, but there may still be a unnecessary rescheduling, so adjust the commit log.

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/pkt_sched.h |  7 +------
 net/core/dev.c          | 26 ++++++++++++++++++++++----
 net/sched/sch_generic.c |  4 +++-
 3 files changed, 26 insertions(+), 11 deletions(-)

include/net/pkt_sched.h +1 -6

index cee1c084e9f4..b16f9236de14 100644
@@ -118,12 +118,7 @@ void __qdisc_run(struct Qdisc *q);
 static inline void qdisc_run(struct Qdisc *q)
 {
 	if (qdisc_run_begin(q)) {
-		/* NOLOCK qdisc must check 'state' under the qdisc seqlock
-		 * to avoid racing with dev_qdisc_reset()
-		 */
-		if (!(q->flags & TCQ_F_NOLOCK) ||
-		    likely(!test_bit(__QDISC_STATE_DEACTIVATED, &q->state)))
-			__qdisc_run(q);
+		__qdisc_run(q);
 		qdisc_run_end(q);
 	}
 }

net/core/dev.c +22 -4

index a30878346f54..0e38b5b044b6 100644
@@ -4515,25 +4515,43 @@ static __latent_entropy void net_tx_action(struct softirq_action *h)
 		sd->output_queue_tailp = &sd->output_queue;
 		local_irq_enable();
 
+		rcu_read_lock();
+
 		while (head) {
 			struct Qdisc *q = head;
 			spinlock_t *root_lock = NULL;
 
 			head = head->next_sched;
 
-			if (!(q->flags & TCQ_F_NOLOCK)) {
-				root_lock = qdisc_lock(q);
-				spin_lock(root_lock);
-			}
 			/* We need to make sure head->next_sched is read
 			 * before clearing __QDISC_STATE_SCHED
 			 */
 			smp_mb__before_atomic();
+
+			if (!(q->flags & TCQ_F_NOLOCK)) {
+				root_lock = qdisc_lock(q);
+				spin_lock(root_lock);
+			} else if (unlikely(test_bit(__QDISC_STATE_DEACTIVATED,
+						     &q->state))) {
+				/* There is a synchronize_net() between
+				 * STATE_DEACTIVATED flag being set and
+				 * qdisc_reset()/some_qdisc_is_busy() in
+				 * dev_deactivate(), so we can safely bail out
+				 * early here to avoid data race between
+				 * qdisc_deactivate() and some_qdisc_is_busy()
+				 * for lockless qdisc.
+				 */
+				clear_bit(__QDISC_STATE_SCHED, &q->state);
+				continue;
+			}
+
 			clear_bit(__QDISC_STATE_SCHED, &q->state);
 			qdisc_run(q);
 			if (root_lock)
 				spin_unlock(root_lock);
 		}
+
+		rcu_read_unlock();
 	}
 
 	xfrm_dev_backlog(sd);

net/sched/sch_generic.c +3 -1

index 0723e7858658..2b87617d023d 100644
@@ -1176,8 +1176,10 @@ static void dev_reset_queue(struct net_device *dev,
 	qdisc_reset(qdisc);
 
 	spin_unlock_bh(qdisc_lock(qdisc));
-	if (nolock)
+	if (nolock) {
+		clear_bit(__QDISC_STATE_MISSED, &qdisc->state);
 		spin_unlock_bh(&qdisc->seqlock);
+	}
 }
 
 static bool some_qdisc_is_busy(struct net_device *dev)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #150 lore raw

PATCH 147/177 net: sched: fix tx action reschedule issue with stopped queue

From: Yunsheng Lin <[email protected]>

[ Upstream commit dcad9ee9e0663d74a89b25b987f9c7be86432812 ]

The netdev qeueue might be stopped when byte queue limit has reached or tx hw ring is full, net_tx_action() may still be rescheduled if STATE_MISSED is set, which consumes unnecessary cpu without dequeuing and transmiting any skb because the netdev queue is stopped, see qdisc_run_end().

This patch fixes it by checking the netdev queue state before calling qdisc_run() and clearing STATE_MISSED if netdev queue is stopped during qdisc_run(), the net_tx_action() is rescheduled again when netdev qeueue is restarted, see netif_tx_wake_queue().

As there is time window between netif_xmit_frozen_or_stopped() checking and STATE_MISSED clearing, between which STATE_MISSED may set by net_tx_action() scheduled by netif_tx_wake_queue(), so set the STATE_MISSED again if netdev queue is restarted.

Fixes 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking")

Reported-by Michal Kubecek <[email protected]>

Acked-by Jakub Kicinski <[email protected]>

Signed-off-by Yunsheng Lin <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/core/dev.c          |  3 ++-
 net/sched/sch_generic.c | 27 ++++++++++++++++++++++++++-
 2 files changed, 28 insertions(+), 2 deletions(-)

net/core/dev.c +2 -1

index 0e38b5b044b6..e226f266da9e 100644
@@ -3384,7 +3384,8 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q,
 
 	if (q->flags & TCQ_F_NOLOCK) {
 		rc = q->enqueue(skb, q, &to_free) & NET_XMIT_MASK;
-		qdisc_run(q);
+		if (likely(!netif_xmit_frozen_or_stopped(txq)))
+			qdisc_run(q);
 
 		if (unlikely(to_free))
 			kfree_skb_list(to_free);

net/sched/sch_generic.c +26 -1

index 2b87617d023d..9bc5cbe9809b 100644
@@ -35,6 +35,25 @@
 const struct Qdisc_ops *default_qdisc_ops = &pfifo_fast_ops;
 EXPORT_SYMBOL(default_qdisc_ops);
 
+static void qdisc_maybe_clear_missed(struct Qdisc *q,
+				     const struct netdev_queue *txq)
+{
+	clear_bit(__QDISC_STATE_MISSED, &q->state);
+
+	/* Make sure the below netif_xmit_frozen_or_stopped()
+	 * checking happens after clearing STATE_MISSED.
+	 */
+	smp_mb__after_atomic();
+
+	/* Checking netif_xmit_frozen_or_stopped() again to
+	 * make sure STATE_MISSED is set if the STATE_MISSED
+	 * set by netif_tx_wake_queue()'s rescheduling of
+	 * net_tx_action() is cleared by the above clear_bit().
+	 */
+	if (!netif_xmit_frozen_or_stopped(txq))
+		set_bit(__QDISC_STATE_MISSED, &q->state);
+}
+
 /* Main transmission queue. */
 
 /* Modifications to data participating in scheduling must be protected with

@@ -74,6 +93,7 @@ static inline struct sk_buff *__skb_dequeue_bad_txq(struct Qdisc *q)
 			}
 		} else {
 			skb = SKB_XOFF_MAGIC;
+			qdisc_maybe_clear_missed(q, txq);
 		}
 	}

@@ -242,6 +262,7 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate,
 			}
 		} else {
 			skb = NULL;
+			qdisc_maybe_clear_missed(q, txq);
 		}
 		if (lock)
 			spin_unlock(lock);

@@ -251,8 +272,10 @@ validate:
 	*validate = true;
 
 	if ((q->flags & TCQ_F_ONETXQUEUE) &&
-	    netif_xmit_frozen_or_stopped(txq))
+	    netif_xmit_frozen_or_stopped(txq)) {
+		qdisc_maybe_clear_missed(q, txq);
 		return skb;
+	}
 
 	skb = qdisc_dequeue_skb_bad_txq(q);
 	if (unlikely(skb)) {

@@ -311,6 +334,8 @@ bool sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q,
 		HARD_TX_LOCK(dev, txq, smp_processor_id());
 		if (!netif_xmit_frozen_or_stopped(txq))
 			skb = dev_hard_start_xmit(skb, dev, txq, &ret);
+		else
+			qdisc_maybe_clear_missed(q, txq);
 
 		HARD_TX_UNLOCK(dev, txq);
 	} else {

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #151 lore raw

PATCH 148/177 net: hso: check for allocation failure in hso_create_bulk_serial_device()

From: Dan Carpenter <[email protected]>

[ Upstream commit 31db0dbd72444abe645d90c20ecb84d668f5af5e ]

In current kernels, small allocations never actually fail so this patch shouldn't affect runtime.

Originally this error handling code written with the idea that if the "serial->tiocmget" allocation failed, then we would continue operating instead of bailing out early. But in later years we added an unchecked dereference on the next line.

	serial->tiocmget->serial_state_notification = kzalloc();
        ^^^^^^^^^^^^^^^^^^

Since these allocations are never going fail in real life, this is mostly a philosophical debate, but I think bailing out early is the correct behavior that the user would want. And generally it's safer to bail as soon an error happens.

Fixes af0de1303c4e ("usb: hso: obey DMA rules in tiocmget")

Reviewed-by Johan Hovold <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/usb/hso.c | 37 ++++++++++++++++++-------------------
 1 file changed, 18 insertions(+), 19 deletions(-)

drivers/net/usb/hso.c +18 ~~-19~~

index 8c2c6d7921f4..0436e1661096 100644
@@ -2619,29 +2619,28 @@ static struct hso_device *hso_create_bulk_serial_device(
 		num_urbs = 2;
 		serial->tiocmget = kzalloc(sizeof(struct hso_tiocmget),
 					   GFP_KERNEL);
+		if (!serial->tiocmget)
+			goto exit;
 		serial->tiocmget->serial_state_notification
 			= kzalloc(sizeof(struct hso_serial_state_notification),
 					   GFP_KERNEL);
-		/* it isn't going to break our heart if serial->tiocmget
-		 *  allocation fails don't bother checking this.
-		 */
-		if (serial->tiocmget && serial->tiocmget->serial_state_notification) {
-			tiocmget = serial->tiocmget;
-			tiocmget->endp = hso_get_ep(interface,
-						    USB_ENDPOINT_XFER_INT,
-						    USB_DIR_IN);
-			if (!tiocmget->endp) {
-				dev_err(&interface->dev, "Failed to find INT IN ep\n");
-				goto exit;
-			}
-
-			tiocmget->urb = usb_alloc_urb(0, GFP_KERNEL);
-			if (tiocmget->urb) {
-				mutex_init(&tiocmget->mutex);
-				init_waitqueue_head(&tiocmget->waitq);
-			} else
-				hso_free_tiomget(serial);
+		if (!serial->tiocmget->serial_state_notification)
+			goto exit;
+		tiocmget = serial->tiocmget;
+		tiocmget->endp = hso_get_ep(interface,
+					    USB_ENDPOINT_XFER_INT,
+					    USB_DIR_IN);
+		if (!tiocmget->endp) {
+			dev_err(&interface->dev, "Failed to find INT IN ep\n");
+			goto exit;
 		}
+
+		tiocmget->urb = usb_alloc_urb(0, GFP_KERNEL);
+		if (tiocmget->urb) {
+			mutex_init(&tiocmget->mutex);
+			init_waitqueue_head(&tiocmget->waitq);
+		} else
+			hso_free_tiomget(serial);
 	}
 	else
 		num_urbs = 1;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #152 lore raw

PATCH 149/177 net: bnx2: Fix error return code in bnx2_init_board()

From: Zhen Lei <[email protected]>

[ Upstream commit 28c66b6da4087b8cfe81c2ec0a46eb6116dafda9 ]

Fix to return -EPERM from the error handling case instead of 0, as done elsewhere in this function.

Fixes b6016b767397 ("[BNX2]: New Broadcom gigabit network driver.")

Reported-by Hulk Robot <[email protected]>

Signed-off-by Zhen Lei <[email protected]>

Reviewed-by Michael Chan <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/broadcom/bnx2.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/ethernet/broadcom/bnx2.c +1 -1

index fbc196b480b6..c3f67d8e1093 100644
@@ -8249,9 +8249,9 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 		BNX2_WR(bp, PCI_COMMAND, reg);
 	} else if ((BNX2_CHIP_ID(bp) == BNX2_CHIP_ID_5706_A1) &&
 		!(bp->flags & BNX2_FLAG_PCIX)) {
-
 		dev_err(&pdev->dev,
 			"5706 A1 can only be used in a PCIX bus, aborting\n");
+		rc = -EPERM;
 		goto err_out_unmap;
 	}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #153 lore raw

PATCH 150/177 bnxt_en: Include new P5 HV definition in VF check.

From: Andy Gospodarek <[email protected]>

[ Upstream commit ab21494be9dc7d62736c5fcd06be65d49df713ee ]

Otherwise, some of the recently added HyperV VF IDs would not be recognized as VF devices and they would not initialize properly.

Fixes 7fbf359bb2c1 ("bnxt_en: Add PCI IDs for Hyper-V VF devices.")

Reviewed-by Edwin Peer <[email protected]>

Signed-off-by Andy Gospodarek <[email protected]>

Signed-off-by Michael Chan <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/net/ethernet/broadcom/bnxt/bnxt.c +2 -1

index 106f2b2ce17f..0dba28bb309a 100644
@@ -280,7 +280,8 @@ static bool bnxt_vf_pciid(enum board_idx idx)
 {
 	return (idx == NETXTREME_C_VF || idx == NETXTREME_E_VF ||
 		idx == NETXTREME_S_VF || idx == NETXTREME_C_VF_HV ||
-		idx == NETXTREME_E_VF_HV || idx == NETXTREME_E_P5_VF);
+		idx == NETXTREME_E_VF_HV || idx == NETXTREME_E_P5_VF ||
+		idx == NETXTREME_E_P5_VF_HV);
 }
 
 #define DB_CP_REARM_FLAGS	(DB_KEY_CP | DB_IDX_VALID)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #154 lore raw

PATCH 151/177 mld: fix panic in mld_newpack()

From: Taehee Yoo <[email protected]>

[ Upstream commit 020ef930b826d21c5446fdc9db80fd72a791bc21 ]

mld_newpack() doesn't allow to allocate high order page, only order-0 allocation is allowed.
If headroom size is too large, a kernel panic could occur in skb_put().

Test commands:
ip netns del A
ip netns del B
ip netns add A
ip netns add B
ip link add veth0 type veth peer name veth1
ip link set veth0 netns A
ip link set veth1 netns B

ip netns exec A ip link set lo up
ip netns exec A ip link set veth0 up
ip netns exec A ip -6 a a 2001:db8:0::1/64 dev veth0 ip netns exec B ip link set lo up
ip netns exec B ip link set veth1 up
ip netns exec B ip -6 a a 2001:db8:0::2/64 dev veth1 for i in {1..99}
do
let A=$i-1
ip netns exec A ip link add ip6gre$i type ip6gre \
local 2001:db8:$A::1 remote 2001:db8:$A::2 encaplimit 100 ip netns exec A ip -6 a a 2001:db8:$i::1/64 dev ip6gre$i ip netns exec A ip link set ip6gre$i up

ip netns exec B ip link add ip6gre$i type ip6gre \
local 2001:db8:$A::2 remote 2001:db8:$A::1 encaplimit 100 ip netns exec B ip -6 a a 2001:db8:$i::2/64 dev ip6gre$i ip netns exec B ip link set ip6gre$i up
done

Splat looks like:
kernel BUG at net/core/skbuff.c:110!
invalid opcode: 0000 [#1] SMP DEBUG_PAGEALLOC KASAN PTI

CPU: 0 PID: 7 Comm: kworker/0:1 Not tainted 5.12.0+ #891
Workqueue: ipv6_addrconf addrconf_dad_work
RIP: 0010:skb_panic+0x15d/0x15f
Code: 92 fe 4c 8b 4c 24 10 53 8b 4d 70 45 89 e0 48 c7 c7 00 ae 79 83
41 57 41 56 41 55 48 8b 54 24 a6 26 f9 ff <0f> 0b 48 8b 6c 24 20 89
34 24 e8 4a 4e 92 fe 8b 34 24 48 c7 c1 20
RSP: 0018:ffff88810091f820 EFLAGS: 00010282
RAX: 0000000000000089 RBX: ffff8881086e9000 RCX: 0000000000000000
RDX: 0000000000000089 RSI: 0000000000000008 RDI: ffffed1020123efb
RBP: ffff888005f6eac0 R08: ffffed1022fc0031 R09: ffffed1022fc0031
R10: ffff888117e00187 R11: ffffed1022fc0030 R12: 0000000000000028
R13: ffff888008284eb0 R14: 0000000000000ed8 R15: 0000000000000ec0
FS:  0000000000000000(0000) GS:ffff888117c00000(0000)
knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f8b801c5640 CR3: 0000000033c2c006 CR4: 00000000003706f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 ? ip6_mc_hdr.isra.26.constprop.46+0x12a/0x600
 ? ip6_mc_hdr.isra.26.constprop.46+0x12a/0x600
 skb_put.cold.104+0x22/0x22
 ip6_mc_hdr.isra.26.constprop.46+0x12a/0x600
 ? rcu_read_lock_sched_held+0x91/0xc0
 mld_newpack+0x398/0x8f0
 ? ip6_mc_hdr.isra.26.constprop.46+0x600/0x600
 ? lock_contended+0xc40/0xc40
 add_grhead.isra.33+0x280/0x380
 add_grec+0x5ca/0xff0
 ? mld_sendpack+0xf40/0xf40
 ? lock_downgrade+0x690/0x690
 mld_send_initial_cr.part.34+0xb9/0x180
 ipv6_mc_dad_complete+0x15d/0x1b0
 addrconf_dad_completed+0x8d2/0xbb0
 ? lock_downgrade+0x690/0x690
 ? addrconf_rs_timer+0x660/0x660
 ? addrconf_dad_work+0x73c/0x10e0
 addrconf_dad_work+0x73c/0x10e0

Allowing high order page allocation could fix this problem.

Fixes 72e09ad107e7 ("ipv6: avoid high order allocations")

Signed-off-by Taehee Yoo <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/ipv6/mcast.c | 3 ---
 1 file changed, 3 deletions(-)

net/ipv6/mcast.c +0 -3

index c875c9b6edbe..7d0a6a7c9d28 100644
@@ -1604,10 +1604,7 @@ static struct sk_buff *mld_newpack(struct inet6_dev *idev, unsigned int mtu)
 		     IPV6_TLV_PADN, 0 };
 
 	/* we assume size > sizeof(ra) here */
-	/* limit our allocations to order-0 page */
-	size = min_t(int, size, SKB_MAX_ORDER(0, 0));
 	skb = sock_alloc_send_skb(sk, size, 1, &err);
-
 	if (!skb)
 		return NULL;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #155 lore raw

PATCH 152/177 gve: Check TX QPL was actually assigned

From: Catherine Sullivan <[email protected]>

[ Upstream commit 5aec55b46c6238506cdf0c60cd0e42ab77a1e5e0 ]

Correctly check the TX QPL was assigned and unassigned if other steps in the allocation fail.

Fixes f5cedc84a30d (gve: Add transmit and receive support)

Signed-off-by Catherine Sullivan <[email protected]>

Acked-by Willem de Bruijn <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/google/gve/gve_tx.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

drivers/net/ethernet/google/gve/gve_tx.c +5 -1

index d0244feb0301..30532ee28dd3 100644
@@ -207,10 +207,12 @@ static int gve_tx_alloc_ring(struct gve_priv *priv, int idx)
 		goto abort_with_info;
 
 	tx->tx_fifo.qpl = gve_assign_tx_qpl(priv);
+	if (!tx->tx_fifo.qpl)
+		goto abort_with_desc;
 
 	/* map Tx FIFO */
 	if (gve_tx_fifo_init(priv, &tx->tx_fifo))
-		goto abort_with_desc;
+		goto abort_with_qpl;
 
 	tx->q_resources =
 		dma_alloc_coherent(hdev,

@@ -229,6 +231,8 @@ static int gve_tx_alloc_ring(struct gve_priv *priv, int idx)
 
 abort_with_fifo:
 	gve_tx_fifo_release(priv, &tx->tx_fifo);
+abort_with_qpl:
+	gve_unassign_qpl(priv, tx->tx_fifo.qpl->id);
 abort_with_desc:
 	dma_free_coherent(hdev, bytes, tx->desc, tx->bus);
 	tx->desc = NULL;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #156 lore raw

PATCH 153/177 gve: Update mgmt_msix_idx if num_ntfy changes

From: David Awogbemila <[email protected]>

[ Upstream commit e96b491a0ffa35a8a9607c193fa4d894ca9fb32f ]

If we do not get the expected number of vectors from pci_enable_msix_range, we update priv->num_ntfy_blks but not priv->mgmt_msix_idx. This patch fixes this so that priv->mgmt_msix_idx is updated accordingly.

Fixes f5cedc84a30d ("gve: Add transmit and receive support")

Acked-by Willem de Bruijn <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/google/gve/gve_main.c | 1 +
 1 file changed, 1 insertion(+)

drivers/net/ethernet/google/gve/gve_main.c +1 -0

index 9b7a8db9860f..9c74251f9b6a 100644
@@ -161,6 +161,7 @@ static int gve_alloc_notify_blocks(struct gve_priv *priv)
 		int vecs_left = new_num_ntfy_blks % 2;
 
 		priv->num_ntfy_blks = new_num_ntfy_blks;
+		priv->mgmt_msix_idx = priv->num_ntfy_blks;
 		priv->tx_cfg.max_queues = min_t(int, priv->tx_cfg.max_queues,
 						vecs_per_type);
 		priv->rx_cfg.max_queues = min_t(int, priv->rx_cfg.max_queues,

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #157 lore raw

PATCH 154/177 gve: Add NULL pointer checks when freeing irqs.

From: David Awogbemila <[email protected]>

[ Upstream commit 5218e919c8d06279884aa0baf76778a6817d5b93 ]

When freeing notification blocks, we index priv->msix_vectors. If we failed to allocate priv->msix_vectors (see abort_with_msix_vectors) this could lead to a NULL pointer dereference if the driver is unloaded.

Fixes 893ce44df565 ("gve: Add basic driver framework for Compute Engine Virtual NIC")

Acked-by Willem de Brujin <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/google/gve/gve_main.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

drivers/net/ethernet/google/gve/gve_main.c +11 -9

index 9c74251f9b6a..77e79d2939ba 100644
@@ -242,20 +242,22 @@ static void gve_free_notify_blocks(struct gve_priv *priv)
 {
 	int i;
 
-	/* Free the irqs */
-	for (i = 0; i < priv->num_ntfy_blks; i++) {
-		struct gve_notify_block *block = &priv->ntfy_blocks[i];
-		int msix_idx = i;
-
-		irq_set_affinity_hint(priv->msix_vectors[msix_idx].vector,
-				      NULL);
-		free_irq(priv->msix_vectors[msix_idx].vector, block);
+	if (priv->msix_vectors) {
+		/* Free the irqs */
+		for (i = 0; i < priv->num_ntfy_blks; i++) {
+			struct gve_notify_block *block = &priv->ntfy_blocks[i];
+			int msix_idx = i;
+
+			irq_set_affinity_hint(priv->msix_vectors[msix_idx].vector,
+					      NULL);
+			free_irq(priv->msix_vectors[msix_idx].vector, block);
+		}
+		free_irq(priv->msix_vectors[priv->mgmt_msix_idx].vector, priv);
 	}
 	dma_free_coherent(&priv->pdev->dev,
 			  priv->num_ntfy_blks * sizeof(*priv->ntfy_blocks),
 			  priv->ntfy_blocks, priv->ntfy_block_bus);
 	priv->ntfy_blocks = NULL;
-	free_irq(priv->msix_vectors[priv->mgmt_msix_idx].vector, priv);
 	pci_disable_msix(priv->pdev);
 	kvfree(priv->msix_vectors);
 	priv->msix_vectors = NULL;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #158 lore raw

PATCH 155/177 gve: Upgrade memory barrier in poll routine

From: Catherine Sullivan <[email protected]>

[ Upstream commit f81781835f0adfae8d701545386030d223efcd6f ]

As currently written, if the driver checks for more work (via gve_tx_poll or gve_rx_poll) before the device posts work and the irq doorbell is not unmasked
(via iowrite32be(GVE_IRQ_ACK | GVE_IRQ_EVENT, ...)) before the device attempts to raise an interrupt, an interrupt is lost and this could potentially lead to the traffic being completely halted. For example, if a tx queue has already been stopped, the driver won't get the chance to complete work and egress will be halted.

We need a full memory barrier in the poll
routine to ensure that the irq doorbell is unmasked before the driver checks for more work.

Fixes f5cedc84a30d ("gve: Add transmit and receive support")

Signed-off-by Catherine Sullivan <[email protected]>

Acked-by Willem de Brujin <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/google/gve/gve_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/ethernet/google/gve/gve_main.c +1 -1

index 77e79d2939ba..6ea0975d74a1 100644
@@ -121,7 +121,7 @@ static int gve_napi_poll(struct napi_struct *napi, int budget)
 	/* Double check we have no extra work.
 	 * Ensure unmask synchronizes with checking for work.
 	 */
-	dma_rmb();
+	mb();
 	if (block->tx)
 		reschedule |= gve_tx_poll(block, -1);
 	if (block->rx)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #159 lore raw

PATCH 156/177 gve: Correct SKB queue index validation.

From: David Awogbemila <[email protected]>

[ Upstream commit fbd4a28b4fa66faaa7f510c0adc531d37e0a7848 ]

SKBs with skb_get_queue_mapping(skb) == tx_cfg.num_queues should also be considered invalid.

Fixes f5cedc84a30d ("gve: Add transmit and receive support")

Acked-by Willem de Brujin <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/google/gve/gve_tx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/ethernet/google/gve/gve_tx.c +1 -1

index 30532ee28dd3..b653197b34d1 100644
@@ -482,7 +482,7 @@ netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev)
 	struct gve_tx_ring *tx;
 	int nsegs;
 
-	WARN(skb_get_queue_mapping(skb) > priv->tx_cfg.num_queues,
+	WARN(skb_get_queue_mapping(skb) >= priv->tx_cfg.num_queues,
 	     "skb queue index out of range");
 	tx = &priv->tx[skb_get_queue_mapping(skb)];
 	if (unlikely(gve_maybe_stop_tx(tx, skb))) {

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #160 lore raw

PATCH 157/177 cxgb4: avoid accessing registers when clearing filters

From: Raju Rangoju <[email protected]>

[ Upstream commit 88c380df84fbd03f9b137c2b9d0a44b9f2f553b0 ]

Hardware register having the server TID base can contain invalid values when adapter is in bad state (for example, due to AER fatal error). Reading these invalid values in the register can lead to out-of-bound memory access. So, fix by using the saved server TID base when clearing filters.

Fixes b1a79360ee86 ("cxgb4: Delete all hash and TCAM filters before resource cleanup")

Signed-off-by Raju Rangoju <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c +1 -1

index 64a2453e06ba..ccb28182f745 100644
@@ -778,7 +778,7 @@ void clear_all_filters(struct adapter *adapter)
 				cxgb4_del_filter(dev, i, &f->fs);
 		}
 
-		sb = t4_read_reg(adapter, LE_DB_SRVR_START_INDEX_A);
+		sb = adapter->tids.stid_base;
 		for (i = 0; i < sb; i++) {
 			f = (struct filter_entry *)adapter->tids.tid_tab[i];

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #161 lore raw

PATCH 158/177 staging: emxx_udc: fix loop in _nbu2ss_nuke()

From: Dan Carpenter <[email protected]>

[ Upstream commit e0112a7c9e847ada15a631b88e279d547e8f26a7 ]

The _nbu2ss_ep_done() function calls:

	list_del_init(&req->queue);

which means that the loop will never exit.

Fixes ca3d253eb967 ("Staging: emxx_udc: Iterate list using list_for_each_entry")

Link https://lore.kernel.org/r/YKUd0sDyjm/lkJfJ@mwanda open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/staging/emxx_udc/emxx_udc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/staging/emxx_udc/emxx_udc.c +2 -2

index a6c893ddbf28..cc4c18c3fb36 100644
@@ -2064,7 +2064,7 @@ static int _nbu2ss_nuke(struct nbu2ss_udc *udc,
 			struct nbu2ss_ep *ep,
 			int status)
 {
-	struct nbu2ss_req *req;
+	struct nbu2ss_req *req, *n;
 
 	/* Endpoint Disable */
 	_nbu2ss_epn_exit(udc, ep);

@@ -2076,7 +2076,7 @@ static int _nbu2ss_nuke(struct nbu2ss_udc *udc,
 		return 0;
 
 	/* called with irqs blocked */
-	list_for_each_entry(req, &ep->queue, queue) {
+	list_for_each_entry_safe(req, n, &ep->queue, queue) {
 		_nbu2ss_ep_done(ep, req, status);
 	}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:31 UTC #162 lore raw

PATCH 159/177 ASoC: cs35l33: fix an error code in probe()

From: Dan Carpenter <[email protected]>

[ Upstream commit 833bc4cf9754643acc69b3c6b65988ca78df4460 ]

This error path returns zero (success) but it should return -EINVAL.

Fixes 3333cb7187b9 ("ASoC: cs35l33: Initial commit of the cs35l33 CODEC driver.")

Link https://lore.kernel.org/r/YKXuyGEzhPT35R3G@mwanda open here

Reviewed-by Charles Keepax <[email protected]>

Signed-off-by Mark Brown <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 sound/soc/codecs/cs35l33.c | 1 +
 1 file changed, 1 insertion(+)

sound/soc/codecs/cs35l33.c +1 -0

index 6042194d95d3..8894369e329a 100644
@@ -1201,6 +1201,7 @@ static int cs35l33_i2c_probe(struct i2c_client *i2c_client,
 		dev_err(&i2c_client->dev,
 			"CS35L33 Device ID (%X). Expected ID %X\n",
 			devid, CS35L33_CHIP_ID);
+		ret = -EINVAL;
 		goto err_enable;
 	}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #163 lore raw

PATCH 160/177 bpf: Set mac_len in bpf_skb_change_head

Link https://lore.kernel.org/bpf/[email protected] open here

From: Jussi Maki <[email protected]>

[ Upstream commit 84316ca4e100d8cbfccd9f774e23817cb2059868 ]

The skb_change_head() helper did not set "skb->mac_len", which is problematic when it's used in combination with skb_redirect_peer(). Without it, redirecting a packet from a L3 device such as wireguard to the veth peer device will cause skb->data to point to the middle of the IP header on entry to tcp_v4_rcv() since the L2 header is not pulled correctly due to mac_len=0.

Fixes 3a0af8fd61f9 ("bpf: BPF for lightweight tunnel infrastructure")

Signed-off-by Jussi Maki <[email protected]>

Signed-off-by Daniel Borkmann <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/core/filter.c | 1 +
 1 file changed, 1 insertion(+)

net/core/filter.c +1 -0

index 7fbb274b7fe3..108bcf600052 100644
@@ -3331,6 +3331,7 @@ static inline int __bpf_skb_change_head(struct sk_buff *skb, u32 head_room,
 		__skb_push(skb, head_room);
 		memset(skb->data, 0, head_room);
 		skb_reset_mac_header(skb);
+		skb_reset_mac_len(skb);
 	}
 
 	return ret;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:30 UTC #164 lore raw

PATCH 161/177 ixgbe: fix large MTU request from VF

From: Jesse Brandeburg <[email protected]>

[ Upstream commit 63e39d29b3da02e901349f6cd71159818a4737a6 ]

Check that the MTU value requested by the VF is in the supported range of MTUs before attempting to set the VF large packet enable, otherwise reject the request. This also avoids unnecessary register updates in the case of the 82599 controller.

Fixes 872844ddb9e4 ("ixgbe: Enable jumbo frames support w/ SR-IOV")

Co-developed-by Piotr Skajewski <[email protected]>

Signed-off-by Piotr Skajewski <[email protected]>

Signed-off-by Jesse Brandeburg <[email protected]>

Co-developed-by Mateusz Palczewski <[email protected]>

Signed-off-by Mateusz Palczewski <[email protected]>

Tested-by Konrad Jankowski <[email protected]>

Signed-off-by Tony Nguyen <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c +7 -9

index 537dfff585e0..47a920128760 100644
@@ -467,12 +467,16 @@ static int ixgbe_set_vf_vlan(struct ixgbe_adapter *adapter, int add, int vid,
 	return err;
 }
 
-static s32 ixgbe_set_vf_lpe(struct ixgbe_adapter *adapter, u32 *msgbuf, u32 vf)
+static int ixgbe_set_vf_lpe(struct ixgbe_adapter *adapter, u32 max_frame, u32 vf)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
-	int max_frame = msgbuf[1];
 	u32 max_frs;
 
+	if (max_frame < ETH_MIN_MTU || max_frame > IXGBE_MAX_JUMBO_FRAME_SIZE) {
+		e_err(drv, "VF max_frame %d out of range\n", max_frame);
+		return -EINVAL;
+	}
+
 	/*
 	 * For 82599EB we have to keep all PFs and VFs operating with
 	 * the same max_frame value in order to avoid sending an oversize

@@ -533,12 +537,6 @@ static s32 ixgbe_set_vf_lpe(struct ixgbe_adapter *adapter, u32 *msgbuf, u32 vf)
 		}
 	}
 
-	/* MTU < 68 is an error and causes problems on some kernels */
-	if (max_frame > IXGBE_MAX_JUMBO_FRAME_SIZE) {
-		e_err(drv, "VF max_frame %d out of range\n", max_frame);
-		return -EINVAL;
-	}
-
 	/* pull current max frame size from hardware */
 	max_frs = IXGBE_READ_REG(hw, IXGBE_MAXFRS);
 	max_frs &= IXGBE_MHADD_MFS_MASK;

@@ -1249,7 +1247,7 @@ static int ixgbe_rcv_msg_from_vf(struct ixgbe_adapter *adapter, u32 vf)
 		retval = ixgbe_set_vf_vlan_msg(adapter, msgbuf, vf);
 		break;
 	case IXGBE_VF_SET_LPE:
-		retval = ixgbe_set_vf_lpe(adapter, msgbuf, vf);
+		retval = ixgbe_set_vf_lpe(adapter, msgbuf[1], vf);
 		break;
 	case IXGBE_VF_SET_MACVLAN:
 		retval = ixgbe_set_vf_macvlan_msg(adapter, msgbuf, vf);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:30 UTC #165 lore raw

PATCH 162/177 scsi: libsas: Use _safe() loop in sas_resume_port()

Link https://lore.kernel.org/r/YKUeq6gwfGcvvhty@mwanda open here

From: Dan Carpenter <[email protected]>

[ Upstream commit 8c7e7b8486cda21269d393245883c5e4737d5ee7 ]

If sas_notify_lldd_dev_found() fails then this code calls:

	sas_unregister_dev(port, dev);

which removes "dev", our list iterator, from the list. This could lead to an endless loop. We need to use list_for_each_entry_safe().

Fixes 303694eeee5e ("[SCSI] libsas: suspend / resume support")

Reviewed-by John Garry <[email protected]>

Signed-off-by Martin K. Petersen <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/scsi/libsas/sas_port.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/scsi/libsas/sas_port.c +2 -2

index 7c86fd248129..f751a12f92ea 100644
@@ -25,7 +25,7 @@ static bool phy_is_wideport_member(struct asd_sas_port *port, struct asd_sas_phy
 
 static void sas_resume_port(struct asd_sas_phy *phy)
 {
-	struct domain_device *dev;
+	struct domain_device *dev, *n;
 	struct asd_sas_port *port = phy->port;
 	struct sas_ha_struct *sas_ha = phy->ha;
 	struct sas_internal *si = to_sas_internal(sas_ha->core.shost->transportt);

@@ -44,7 +44,7 @@ static void sas_resume_port(struct asd_sas_phy *phy)
 	 * 1/ presume every device came back
 	 * 2/ force the next revalidation to check all expander phys
 	 */
-	list_for_each_entry(dev, &port->dev_list, dev_list_node) {
+	list_for_each_entry_safe(dev, n, &port->dev_list, dev_list_node) {
 		int i, rc;
 
 		rc = sas_notify_lldd_dev_found(dev);

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:30 UTC #166 lore raw

PATCH 163/177 net: lantiq: fix memory corruption in RX ring

From: Aleksander Jan Bajkowski <[email protected]>

[ Upstream commit c7718ee96dbc2f9c5fc3b578abdf296dd44b9c20 ]

In a situation where memory allocation or dma mapping fails, an invalid address is programmed into the descriptor. This can lead to memory corruption. If the memory allocation fails, DMA should reuse the previous skb and mapping and drop the packet. This patch also increments rx drop counter.

Fixes fe1a56420cf2 ("net: lantiq: Add Lantiq / Intel VRX200 Ethernet driver ")

Signed-off-by Aleksander Jan Bajkowski <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/lantiq_xrx200.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

drivers/net/ethernet/lantiq_xrx200.c +9 -5

index 4e44a39267eb..6ece99e6b6dd 100644
@@ -154,6 +154,7 @@ static int xrx200_close(struct net_device *net_dev)
 
 static int xrx200_alloc_skb(struct xrx200_chan *ch)
 {
+	dma_addr_t mapping;
 	int ret = 0;
 
 	ch->skb[ch->dma.desc] = netdev_alloc_skb_ip_align(ch->priv->net_dev,

@@ -163,16 +164,17 @@ static int xrx200_alloc_skb(struct xrx200_chan *ch)
 		goto skip;
 	}
 
-	ch->dma.desc_base[ch->dma.desc].addr = dma_map_single(ch->priv->dev,
-			ch->skb[ch->dma.desc]->data, XRX200_DMA_DATA_LEN,
-			DMA_FROM_DEVICE);
-	if (unlikely(dma_mapping_error(ch->priv->dev,
-				       ch->dma.desc_base[ch->dma.desc].addr))) {
+	mapping = dma_map_single(ch->priv->dev, ch->skb[ch->dma.desc]->data,
+				 XRX200_DMA_DATA_LEN, DMA_FROM_DEVICE);
+	if (unlikely(dma_mapping_error(ch->priv->dev, mapping))) {
 		dev_kfree_skb_any(ch->skb[ch->dma.desc]);
 		ret = -ENOMEM;
 		goto skip;
 	}
 
+	ch->dma.desc_base[ch->dma.desc].addr = mapping;
+	/* Make sure the address is written before we give it to HW */
+	wmb();
 skip:
 	ch->dma.desc_base[ch->dma.desc].ctl =
 		LTQ_DMA_OWN | LTQ_DMA_RX_OFFSET(NET_IP_ALIGN) |

@@ -196,6 +198,8 @@ static int xrx200_hw_receive(struct xrx200_chan *ch)
 	ch->dma.desc %= LTQ_DESC_NUM;
 
 	if (ret) {
+		ch->skb[ch->dma.desc] = skb;
+		net_dev->stats.rx_dropped++;
 		netdev_err(net_dev, "failed to allocate new rx buffer\n");
 		return ret;
 	}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:30 UTC #167 lore raw

PATCH 164/177 ipv6: record frag_max_size in atomic fragments in input path

From: Francesco Ruggeri <[email protected]>

[ Upstream commit e29f011e8fc04b2cdc742a2b9bbfa1b62518381a ]

Commit dbd1759e6a9c ("ipv6: on reassembly, record frag_max_size") filled the frag_max_size field in IP6CB in the input path. The field should also be filled in case of atomic fragments.

Fixes dbd1759e6a9c ('ipv6: on reassembly, record frag_max_size')

Signed-off-by Francesco Ruggeri <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/ipv6/reassembly.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

net/ipv6/reassembly.c +3 -1

index c8cf1bbad74a..45ee1971d998 100644
@@ -344,7 +344,7 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 	hdr = ipv6_hdr(skb);
 	fhdr = (struct frag_hdr *)skb_transport_header(skb);
 
-	if (!(fhdr->frag_off & htons(0xFFF9))) {
+	if (!(fhdr->frag_off & htons(IP6_OFFSET | IP6_MF))) {
 		/* It is not a fragmented frame */
 		skb->transport_header += sizeof(struct frag_hdr);
 		__IP6_INC_STATS(net,

@@ -352,6 +352,8 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 
 		IP6CB(skb)->nhoff = (u8 *)fhdr - skb_network_header(skb);
 		IP6CB(skb)->flags |= IP6SKB_FRAGMENTED;
+		IP6CB(skb)->frag_max_size = ntohs(hdr->payload_len) +
+					    sizeof(struct ipv6hdr);
 		return 1;
 	}

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:30 UTC #168 lore raw

PATCH 165/177 ALSA: usb-audio: scarlett2: snd_scarlett_gen2_controls_create() can be static

Link https://lore.kernel.org/r/20210522180900.GA83915@f59a3af2f1d9 open here

From: kernel test robot <[email protected]>

[ Upstream commit 2b899f31f1a6db2db4608bac2ac04fe2c4ad89eb ]

sound/usb/mixer_scarlett_gen2.c:2000:5: warning: symbol 'snd_scarlett_gen2_controls_create' was not declared. Should it be static?

Fixes 265d1a90e4fb ("ALSA: usb-audio: scarlett2: Improve driver startup messages")

Reported-by kernel test robot <[email protected]>

Signed-off-by kernel test robot <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 sound/usb/mixer_scarlett_gen2.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

sound/usb/mixer_scarlett_gen2.c +2 -2

index a973dd4d5bbe..7a10c9e22c46 100644
@@ -1997,8 +1997,8 @@ static int scarlett2_mixer_status_create(struct usb_mixer_interface *mixer)
 	return usb_submit_urb(mixer->urb, GFP_KERNEL);
 }
 
-int snd_scarlett_gen2_controls_create(struct usb_mixer_interface *mixer,
-				      const struct scarlett2_device_info *info)
+static int snd_scarlett_gen2_controls_create(struct usb_mixer_interface *mixer,
+					     const struct scarlett2_device_info *info)
 {
 	int err;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:31 UTC #169 lore raw

PATCH 166/177 net: ethernet: mtk_eth_soc: Fix packet statistics support for MT7628/88

From: Stefan Roese <[email protected]>

[ Upstream commit ad79fd2c42f7626bdf6935cd72134c2a5a59ff2d ]

The MT7628/88 SoC(s) have other (limited) packet counter registers than currently supported in the mtk_eth_soc driver. This patch adds support for reading these registers, so that the packet statistics are correctly updated.

Additionally the defines for the non-MT7628 variant packet counter registers are added and used in this patch instead of using hard coded values.

Signed-off-by Stefan Roese <[email protected]>

Fixes 296c9120752b ("net: ethernet: mediatek: Add MT7628/88 SoC support")

Cc Felix Fietkau <[email protected]>

Cc John Crispin <[email protected]>

Cc Ilya Lipnitskiy <[email protected]>

Cc Reto Schneider <[email protected]>

Cc David S. Miller <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 67 ++++++++++++++-------
 drivers/net/ethernet/mediatek/mtk_eth_soc.h | 24 +++++++-
 2 files changed, 66 insertions(+), 25 deletions(-)

drivers/net/ethernet/mediatek/mtk_eth_soc.c +44 ~~-23~~

index 7e3806fd70b2..48b395b9c15a 100644
@@ -675,32 +675,53 @@ static int mtk_set_mac_address(struct net_device *dev, void *p)
 void mtk_stats_update_mac(struct mtk_mac *mac)
 {
 	struct mtk_hw_stats *hw_stats = mac->hw_stats;
-	unsigned int base = MTK_GDM1_TX_GBCNT;
-	u64 stats;
-
-	base += hw_stats->reg_offset;
+	struct mtk_eth *eth = mac->hw;
 
 	u64_stats_update_begin(&hw_stats->syncp);
 
-	hw_stats->rx_bytes += mtk_r32(mac->hw, base);
-	stats =  mtk_r32(mac->hw, base + 0x04);
-	if (stats)
-		hw_stats->rx_bytes += (stats << 32);
-	hw_stats->rx_packets += mtk_r32(mac->hw, base + 0x08);
-	hw_stats->rx_overflow += mtk_r32(mac->hw, base + 0x10);
-	hw_stats->rx_fcs_errors += mtk_r32(mac->hw, base + 0x14);
-	hw_stats->rx_short_errors += mtk_r32(mac->hw, base + 0x18);
-	hw_stats->rx_long_errors += mtk_r32(mac->hw, base + 0x1c);
-	hw_stats->rx_checksum_errors += mtk_r32(mac->hw, base + 0x20);
-	hw_stats->rx_flow_control_packets +=
-					mtk_r32(mac->hw, base + 0x24);
-	hw_stats->tx_skip += mtk_r32(mac->hw, base + 0x28);
-	hw_stats->tx_collisions += mtk_r32(mac->hw, base + 0x2c);
-	hw_stats->tx_bytes += mtk_r32(mac->hw, base + 0x30);
-	stats =  mtk_r32(mac->hw, base + 0x34);
-	if (stats)
-		hw_stats->tx_bytes += (stats << 32);
-	hw_stats->tx_packets += mtk_r32(mac->hw, base + 0x38);
+	if (MTK_HAS_CAPS(eth->soc->caps, MTK_SOC_MT7628)) {
+		hw_stats->tx_packets += mtk_r32(mac->hw, MT7628_SDM_TPCNT);
+		hw_stats->tx_bytes += mtk_r32(mac->hw, MT7628_SDM_TBCNT);
+		hw_stats->rx_packets += mtk_r32(mac->hw, MT7628_SDM_RPCNT);
+		hw_stats->rx_bytes += mtk_r32(mac->hw, MT7628_SDM_RBCNT);
+		hw_stats->rx_checksum_errors +=
+			mtk_r32(mac->hw, MT7628_SDM_CS_ERR);
+	} else {
+		unsigned int offs = hw_stats->reg_offset;
+		u64 stats;
+
+		hw_stats->rx_bytes += mtk_r32(mac->hw,
+					      MTK_GDM1_RX_GBCNT_L + offs);
+		stats = mtk_r32(mac->hw, MTK_GDM1_RX_GBCNT_H + offs);
+		if (stats)
+			hw_stats->rx_bytes += (stats << 32);
+		hw_stats->rx_packets +=
+			mtk_r32(mac->hw, MTK_GDM1_RX_GPCNT + offs);
+		hw_stats->rx_overflow +=
+			mtk_r32(mac->hw, MTK_GDM1_RX_OERCNT + offs);
+		hw_stats->rx_fcs_errors +=
+			mtk_r32(mac->hw, MTK_GDM1_RX_FERCNT + offs);
+		hw_stats->rx_short_errors +=
+			mtk_r32(mac->hw, MTK_GDM1_RX_SERCNT + offs);
+		hw_stats->rx_long_errors +=
+			mtk_r32(mac->hw, MTK_GDM1_RX_LENCNT + offs);
+		hw_stats->rx_checksum_errors +=
+			mtk_r32(mac->hw, MTK_GDM1_RX_CERCNT + offs);
+		hw_stats->rx_flow_control_packets +=
+			mtk_r32(mac->hw, MTK_GDM1_RX_FCCNT + offs);
+		hw_stats->tx_skip +=
+			mtk_r32(mac->hw, MTK_GDM1_TX_SKIPCNT + offs);
+		hw_stats->tx_collisions +=
+			mtk_r32(mac->hw, MTK_GDM1_TX_COLCNT + offs);
+		hw_stats->tx_bytes +=
+			mtk_r32(mac->hw, MTK_GDM1_TX_GBCNT_L + offs);
+		stats =  mtk_r32(mac->hw, MTK_GDM1_TX_GBCNT_H + offs);
+		if (stats)
+			hw_stats->tx_bytes += (stats << 32);
+		hw_stats->tx_packets +=
+			mtk_r32(mac->hw, MTK_GDM1_TX_GPCNT + offs);
+	}
+
 	u64_stats_update_end(&hw_stats->syncp);
 }

drivers/net/ethernet/mediatek/mtk_eth_soc.h +22 -2

index 1e9202b34d35..c0b2768b480f 100644
@@ -264,8 +264,21 @@
 /* QDMA FQ Free Page Buffer Length Register */
 #define MTK_QDMA_FQ_BLEN	0x1B2C
 
-/* GMA1 Received Good Byte Count Register */
-#define MTK_GDM1_TX_GBCNT	0x2400
+/* GMA1 counter / statics register */
+#define MTK_GDM1_RX_GBCNT_L	0x2400
+#define MTK_GDM1_RX_GBCNT_H	0x2404
+#define MTK_GDM1_RX_GPCNT	0x2408
+#define MTK_GDM1_RX_OERCNT	0x2410
+#define MTK_GDM1_RX_FERCNT	0x2414
+#define MTK_GDM1_RX_SERCNT	0x2418
+#define MTK_GDM1_RX_LENCNT	0x241c
+#define MTK_GDM1_RX_CERCNT	0x2420
+#define MTK_GDM1_RX_FCCNT	0x2424
+#define MTK_GDM1_TX_SKIPCNT	0x2428
+#define MTK_GDM1_TX_COLCNT	0x242c
+#define MTK_GDM1_TX_GBCNT_L	0x2430
+#define MTK_GDM1_TX_GBCNT_H	0x2434
+#define MTK_GDM1_TX_GPCNT	0x2438
 #define MTK_STAT_OFFSET		0x40
 
 /* QDMA descriptor txd4 */

@@ -476,6 +489,13 @@
 #define MT7628_SDM_MAC_ADRL	(MT7628_SDM_OFFSET + 0x0c)
 #define MT7628_SDM_MAC_ADRH	(MT7628_SDM_OFFSET + 0x10)
 
+/* Counter / stat register */
+#define MT7628_SDM_TPCNT	(MT7628_SDM_OFFSET + 0x100)
+#define MT7628_SDM_TBCNT	(MT7628_SDM_OFFSET + 0x104)
+#define MT7628_SDM_RPCNT	(MT7628_SDM_OFFSET + 0x108)
+#define MT7628_SDM_RBCNT	(MT7628_SDM_OFFSET + 0x10c)
+#define MT7628_SDM_CS_ERR	(MT7628_SDM_OFFSET + 0x110)
+
 struct mtk_rx_dma {
 	unsigned int rxd1;
 	unsigned int rxd2;

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:31 UTC #170 lore raw

PATCH 167/177 sch_dsmark: fix a NULL deref in qdisc_reset()

From: Taehee Yoo <[email protected]>

[ Upstream commit 9b76eade16423ef06829cccfe3e100cfce31afcd ]

If Qdisc_ops->init() is failed, Qdisc_ops->reset() would be called. When dsmark_init(Qdisc_ops->init()) is failed, it possibly doesn't initialize dsmark_qdisc_data->q. But dsmark_reset(Qdisc_ops->reset()) uses dsmark_qdisc_data->q pointer wihtout any null checking. So, panic would occur.

Test commands:
sysctl net.core.default_qdisc=dsmark -w
ip link add dummy0 type dummy
ip link add vw0 link dummy0 type virt_wifi
ip link set vw0 up

Splat looks like:

KASAN: null-ptr-deref in range [0x0000000000000018-0x000000000000001f]
CPU: 3 PID: 684 Comm: ip Not tainted 5.12.0+ #910
RIP: 0010:qdisc_reset+0x2b/0x680
Code: 1f 44 00 00 48 b8 00 00 00 00 00 fc ff df 41 57 41 56 41 55 41 54
55 48 89 fd 48 83 c7 18 53 48 89 fa 48 c1 ea 03 48 83 ec 20 <80> 3c 02
00 0f 85 09 06 00 00 4c 8b 65 18 0f 1f 44 00 00 65 8b 1d
RSP: 0018:ffff88800fda6bf8 EFLAGS: 00010282
RAX: dffffc0000000000 RBX: ffff8880050ed800 RCX: 0000000000000000
RDX: 0000000000000003 RSI: ffffffff99e34100 RDI: 0000000000000018
RBP: 0000000000000000 R08: fffffbfff346b553 R09: fffffbfff346b553
R10: 0000000000000001 R11: fffffbfff346b552 R12: ffffffffc0824940
R13: ffff888109e83800 R14: 00000000ffffffff R15: ffffffffc08249e0
FS:  00007f5042287680(0000) GS:ffff888119800000(0000)
knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000055ae1f4dbd90 CR3: 0000000006760002 CR4: 00000000003706e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 ? rcu_read_lock_bh_held+0xa0/0xa0
 dsmark_reset+0x3d/0xf0 [sch_dsmark]
 qdisc_reset+0xa9/0x680
 qdisc_destroy+0x84/0x370
 qdisc_create_dflt+0x1fe/0x380
 attach_one_default_qdisc.constprop.41+0xa4/0x180
 dev_activate+0x4d5/0x8c0
 ? __dev_open+0x268/0x390
 __dev_open+0x270/0x390

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")

Signed-off-by Taehee Yoo <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/sched/sch_dsmark.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

net/sched/sch_dsmark.c +2 -1

index 2b88710994d7..76ed1a05ded2 100644
@@ -406,7 +406,8 @@ static void dsmark_reset(struct Qdisc *sch)
 	struct dsmark_qdisc_data *p = qdisc_priv(sch);
 
 	pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p);
-	qdisc_reset(p->q);
+	if (p->q)
+		qdisc_reset(p->q);
 	sch->qstats.backlog = 0;
 	sch->q.qlen = 0;
 }

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:31 UTC #171 lore raw

PATCH 168/177 MIPS: alchemy: xxs1500: add gpio-au1000.h header file

From: Randy Dunlap <[email protected]>

[ Upstream commit ff4cff962a7eedc73e54b5096693da7f86c61346 ]

board-xxs1500.c references 2 functions without declaring them, so add the header file to placate the build.

../arch/mips/alchemy/board-xxs1500.c: In function 'board_setup':
../arch/mips/alchemy/board-xxs1500.c:56:2: error: implicit declaration of function 'alchemy_gpio1_input_enable' [-Werror=implicit-function-declaration]
56 | alchemy_gpio1_input_enable();
../arch/mips/alchemy/board-xxs1500.c:57:2: error: implicit declaration of function 'alchemy_gpio2_enable'; did you mean 'alchemy_uart_enable'? [-Werror=implicit-function-declaration]
57 | alchemy_gpio2_enable();

Fixes 8e026910fcd4 ("MIPS: Alchemy: merge GPR/MTX-1/XXS1500 board code into single files")

Signed-off-by Randy Dunlap <[email protected]>

Cc Thomas Bogendoerfer <[email protected]>

Cc Manuel Lauss <[email protected]>

Cc Ralf Baechle <[email protected]>

Acked-by Manuel Lauss <[email protected]>

Signed-off-by Thomas Bogendoerfer <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/mips/alchemy/board-xxs1500.c | 1 +
 1 file changed, 1 insertion(+)

arch/mips/alchemy/board-xxs1500.c +1 -0

index c67dfe1f4997..ec35aedc7727 100644
@@ -18,6 +18,7 @@
 #include <asm/reboot.h>
 #include <asm/setup.h>
 #include <asm/mach-au1x00/au1000.h>
+#include <asm/mach-au1x00/gpio-au1000.h>
 #include <prom.h>
 
 const char *get_system_type(void)

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #172 lore raw

PATCH 169/177 MIPS: ralink: export rt_sysc_membase for rt2880_wdt.c

From: Randy Dunlap <[email protected]>

[ Upstream commit fef532ea0cd871afab7d9a7b6e9da99ac2c24371 ]

rt2880_wdt.c uses (well, attempts to use) rt_sysc_membase. However, when this watchdog driver is built as a loadable module, there is a build error since the rt_sysc_membase symbol is not exported. Export it to quell the build error.

ERROR: modpost: "rt_sysc_membase" [drivers/watchdog/rt2880_wdt.ko] undefined!

Fixes 473cf939ff34 ("watchdog: add ralink watchdog driver")

Signed-off-by Randy Dunlap <[email protected]>

Cc Guenter Roeck <[email protected]>

Cc Wim Van Sebroeck <[email protected]>

Cc John Crispin <[email protected]>

Acked-by Guenter Roeck <[email protected]>

Signed-off-by Thomas Bogendoerfer <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/mips/ralink/of.c | 2 ++
 1 file changed, 2 insertions(+)

arch/mips/ralink/of.c +2 -0

index 59b23095bfbb..4e38a905ab38 100644
@@ -8,6 +8,7 @@
 
 #include <linux/io.h>
 #include <linux/clk.h>
+#include <linux/export.h>
 #include <linux/init.h>
 #include <linux/sizes.h>
 #include <linux/of_fdt.h>

@@ -25,6 +26,7 @@
 
 __iomem void *rt_sysc_membase;
 __iomem void *rt_memc_membase;
+EXPORT_SYMBOL_GPL(rt_sysc_membase);
 
 __iomem void *plat_of_remap_node(const char *node)
 {

Signature · 1 line

2.30.2

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #173 lore raw

PATCH 170/177 drm/i915/display: fix compiler warning about array overrun

From: Linus Torvalds <[email protected]>

commit fec4d42724a1bf3dcba52307e55375fdb967b852 upstream.

intel_dp_check_mst_status() uses a 14-byte array to read the DPRX Event Status Indicator data, but then passes that buffer at offset 10 off as an argument to drm_dp_channel_eq_ok().

End result: there are only 4 bytes remaining of the buffer, yet drm_dp_channel_eq_ok() wants a 6-byte buffer. gcc-11 correctly warns about this case:

drivers/gpu/drm/i915/display/intel_dp.c: In function ‘intel_dp_check_mst_status’:
drivers/gpu/drm/i915/display/intel_dp.c:3491:22: warning: ‘drm_dp_channel_eq_ok’ reading 6 bytes from a region of size 4 [-Wstringop-overread]

   3491 |                     !drm_dp_channel_eq_ok(&esi[10], intel_dp->lane_count)) {
        |                      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  drivers/gpu/drm/i915/display/intel_dp.c:3491:22: note: referencing argument 1 of type ‘const u8 *’ {aka ‘const unsigned char *’}
  In file included from drivers/gpu/drm/i915/display/intel_dp.c:38:
  include/drm/drm_dp_helper.h:1466:6: note: in a call to function ‘drm_dp_channel_eq_ok’
   1466 | bool drm_dp_channel_eq_ok(const u8 link_status[DP_LINK_STATUS_SIZE],
        |      ^~~~~~~~~~~~~~~~~~~~
       6:14 elapsed

This commit just extends the original array by 2 zero-initialized bytes, avoiding the warning.

There may be some underlying bug in here that caused this confusion, but this is at least no worse than the existing situation that could use random data off the stack.

Cc Jani Nikula <[email protected]>

Cc Ville Syrjälä <[email protected]>

Cc Joonas Lahtinen <[email protected]>

Cc Rodrigo Vivi <[email protected]>

Cc Daniel Vetter <[email protected]>

Cc Dave Airlie <[email protected]>

Signed-off-by Linus Torvalds <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/gpu/drm/i915/display/intel_dp.c |   13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

drivers/gpu/drm/i915/display/intel_dp.c +12 -1

@@ -4706,7 +4706,18 @@ intel_dp_check_mst_status(struct intel_d
 	bool bret;
 
 	if (intel_dp->is_mst) {
-		u8 esi[DP_DPRX_ESI_LEN] = { 0 };
+		/*
+		 * The +2 is because DP_DPRX_ESI_LEN is 14, but we then
+		 * pass in "esi+10" to drm_dp_channel_eq_ok(), which
+		 * takes a 6-byte array. So we actually need 16 bytes
+		 * here.
+		 *
+		 * Somebody who knows what the limits actually are
+		 * should check this, but for now this is at least
+		 * harmless and avoids a valid compiler warning about
+		 * using more of the array than we have allocated.
+		 */
+		u8 esi[DP_DPRX_ESI_LEN+2] = {};
 		int ret = 0;
 		int retry;
 		bool handled;

Greg Kroah-Hartman

author maintainer

2021-05-31 14:29 UTC #174 lore raw

PATCH 171/177 i915: fix build warning in intel_dp_get_link_status()