Skip to contentSkip to conversation

4.14.99-stable review

PATCH In review Greg Kroah-Hartman 8 participants 215 messages Feb 11, 2019 – Feb 13, 2019 Page 2 of 2

Subsystems (6)
Stable branch KVM MIPS s390 +2 more subsystems
lore
Share & tools
lore
author maintainer
#174 lore raw

PATCH 174/205 net: dsa: slave: Dont propagate flag changes on down slave interfaces

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rundong Ge <[email protected]>

[ Upstream commit 17ab4f61b8cd6f9c38e9d0b935d86d73b5d0d2b5 ]

The unbalance of master's promiscuity or allmulti will happen after ifdown and ifup a slave interface which is in a bridge.

When we ifdown a slave interface , both the 'dsa_slave_close' and 'dsa_slave_change_rx_flags' will clear the master's flags. The flags of master will be decrease twice.
In the other hand, if we ifup the slave interface again, since the slave's flags were cleared the 'dsa_slave_open' won't set the master's flag, only 'dsa_slave_change_rx_flags' that triggered by 'br_add_if' will set the master's flags. The flags of master is increase once.

Only propagating flag changes when a slave interface is up makes sure this does not happen. The 'vlan_dev_change_rx_flags' had the same problem and was fixed, and changes here follows that fix.

Fixes 91da11f870f0 ("net: Distributed Switch Architecture protocol support")
---
 net/dsa/slave.c |   12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
net/dsa/slave.c +8 -4
@@ -156,10 +156,14 @@ static void dsa_slave_change_rx_flags(st
 	struct dsa_slave_priv *p = netdev_priv(dev);
 	struct net_device *master = dsa_master_netdev(p);
 
-	if (change & IFF_ALLMULTI)
-		dev_set_allmulti(master, dev->flags & IFF_ALLMULTI ? 1 : -1);
-	if (change & IFF_PROMISC)
-		dev_set_promiscuity(master, dev->flags & IFF_PROMISC ? 1 : -1);
+	if (dev->flags & IFF_UP) {
+		if (change & IFF_ALLMULTI)
+			dev_set_allmulti(master,
+					 dev->flags & IFF_ALLMULTI ? 1 : -1);
+		if (change & IFF_PROMISC)
+			dev_set_promiscuity(master,
+					    dev->flags & IFF_PROMISC ? 1 : -1);
+	}
 }
 
 static void dsa_slave_set_rx_mode(struct net_device *dev)
author maintainer
#175 lore raw

PATCH 175/205 ALSA: compress: Fix stop handling on compressed capture streams

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Charles Keepax <[email protected]>

commit 4f2ab5e1d13d6aa77c55f4914659784efd776eb4 upstream.

It is normal user behaviour to start, stop, then start a stream again without closing it. Currently this works for compressed playback streams but not capture ones.

The states on a compressed capture stream go directly from OPEN to PREPARED, unlike a playback stream which moves to SETUP and waits for a write of data before moving to PREPARED. Currently however, when a stop is sent the state is set to SETUP for both types of streams. This leaves a capture stream in the situation where a new start can't be sent as that requires the state to be PREPARED and a new set_params can't be sent as that requires the state to be OPEN. The only option being to close the stream, and then reopen.

Correct this issues by allowing snd_compr_drain_notify to set the state depending on the stream direction, as we already do in set_params.

Fixes 49bb6402f1aa ("ALSA: compress_core: Add support for capture streams")
---
 include/sound/compress_driver.h |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
include/sound/compress_driver.h +5 -1
@@ -186,7 +186,11 @@ static inline void snd_compr_drain_notif
 	if (snd_BUG_ON(!stream))
 		return;
 
-	stream->runtime->state = SNDRV_PCM_STATE_SETUP;
+	if (stream->direction == SND_COMPRESS_PLAYBACK)
+		stream->runtime->state = SNDRV_PCM_STATE_SETUP;
+	else
+		stream->runtime->state = SNDRV_PCM_STATE_PREPARED;
+
 	wake_up(&stream->runtime->sleep);
 }
 
author maintainer
#176 lore raw

PATCH 176/205 ALSA: hda - Serialize codec registrations

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Takashi Iwai <[email protected]>

commit 305a0ade180981686eec1f92aa6252a7c6ebb1cf upstream.

In the current code, the codec registration may happen both at the codec bind time and the end of the controller probe time. In a rare occasion, they race with each other, leading to Oops due to the still uninitialized card device.

This patch introduces a simple flag to prevent the codec registration at the codec bind time as long as the controller probe is going on. The controller probe invokes snd_card_register() that does the whole registration task, and we don't need to register each piece beforehand.

---
 sound/pci/hda/hda_bind.c  |    3 ++-
 sound/pci/hda/hda_codec.h |    1 +
 sound/pci/hda/hda_intel.c |    2 ++
 3 files changed, 5 insertions(+), 1 deletion(-)
sound/pci/hda/hda_bind.c +2 -1
@@ -109,7 +109,8 @@ static int hda_codec_driver_probe(struct
 	err = snd_hda_codec_build_controls(codec);
 	if (err < 0)
 		goto error_module;
-	if (codec->card->registered) {
+	/* only register after the bus probe finished; otherwise it's racy */
+	if (!codec->bus->bus_probing && codec->card->registered) {
 		err = snd_card_register(codec->card);
 		if (err < 0)
 			goto error_module;
sound/pci/hda/hda_codec.h +1 -0
@@ -68,6 +68,7 @@ struct hda_bus {
 	unsigned int response_reset:1;	/* controller was reset */
 	unsigned int in_reset:1;	/* during reset operation */
 	unsigned int no_response_fallback:1; /* don't fallback at RIRB error */
+	unsigned int bus_probing :1;	/* during probing process */
 
 	int primary_dig_out_type;	/* primary digital out PCM type */
 	unsigned int mixer_assigned;	/* codec addr for mixer name */
sound/pci/hda/hda_intel.c +2 -0
@@ -2236,6 +2236,7 @@ static int azx_probe_continue(struct azx
 	int val;
 	int err;
 
+	to_hda_bus(bus)->bus_probing = 1;
 	hda->probe_continued = 1;
 
 	/* bind with i915 if needed */
@@ -2341,6 +2342,7 @@ i915_power_fail:
 	if (err < 0)
 		hda->init_failed = 1;
 	complete_all(&hda->probe_wait);
+	to_hda_bus(bus)->bus_probing = 0;
 	return err;
 }
 
author maintainer
#177 lore raw

PATCH 177/205 fuse: call pipe_buf_release() under pipe lock

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jann Horn <[email protected]>

commit 9509941e9c534920ccc4771ae70bd6cbbe79df1c upstream.

Some of the pipe_buf_release() handlers seem to assume that the pipe is locked - in particular, anon_pipe_buf_release() accesses pipe->tmp_page without taking any extra locks. From a glance through the callers of pipe_buf_release(), it looks like FUSE is the only one that calls pipe_buf_release() without having the pipe locked.

This bug should only lead to a memory leak, nothing terrible.

Fixes dd3bb14f44a6 ("fuse: support splice() writing to fuse device")
---
 fs/fuse/dev.c |    2 ++
 1 file changed, 2 insertions(+)
fs/fuse/dev.c +2 -0
@@ -2024,8 +2024,10 @@ static ssize_t fuse_dev_splice_write(str
 
 	ret = fuse_dev_do_write(fud, &cs, len);
 
+	pipe_lock(pipe);
 	for (idx = 0; idx < nbuf; idx++)
 		pipe_buf_release(pipe, &bufs[idx]);
+	pipe_unlock(pipe);
 
 out:
 	kfree(bufs);
author maintainer
#178 lore raw

PATCH 178/205 fuse: decrement NR_WRITEBACK_TEMP on the right page

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miklos Szeredi <[email protected]>

commit a2ebba824106dabe79937a9f29a875f837e1b6d4 upstream.

NR_WRITEBACK_TEMP is accounted on the temporary page in the request, not the page cache page.

Fixes 8b284dc47291 ("fuse: writepages: handle same page rewrites")
Cc <[email protected]> # v3.13
---
 fs/fuse/file.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/fuse/file.c +1 -1
@@ -1777,7 +1777,7 @@ static bool fuse_writepage_in_flight(str
 		spin_unlock(&fc->lock);
 
 		dec_wb_stat(&bdi->wb, WB_WRITEBACK);
-		dec_node_page_state(page, NR_WRITEBACK_TEMP);
+		dec_node_page_state(new_req->pages[0], NR_WRITEBACK_TEMP);
 		wb_writeout_inc(&bdi->wb);
 		fuse_writepage_free(fc, new_req);
 		fuse_request_free(new_req);
author maintainer
#179 lore raw

PATCH 179/205 fuse: handle zero sized retrieve correctly

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miklos Szeredi <[email protected]>

commit 97e1532ef81acb31c30f9e75bf00306c33a77812 upstream.

Dereferencing req->page_descs[0] will Oops if req->max_pages is zero.

Reported-by [email protected]
Fixes b2430d7567a3 ("fuse: add per-page descriptor <offset, length> to fuse_req")
Cc <[email protected]> # v3.9
---
 fs/fuse/dev.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/fuse/dev.c +1 -1
@@ -1691,7 +1691,6 @@ static int fuse_retrieve(struct fuse_con
 	req->in.h.nodeid = outarg->nodeid;
 	req->in.numargs = 2;
 	req->in.argpages = 1;
-	req->page_descs[0].offset = offset;
 	req->end = fuse_retrieve_end;
 
 	index = outarg->offset >> PAGE_SHIFT;
@@ -1706,6 +1705,7 @@ static int fuse_retrieve(struct fuse_con
 
 		this_num = min_t(unsigned, num, PAGE_SIZE - offset);
 		req->pages[req->num_pages] = page;
+		req->page_descs[req->num_pages].offset = offset;
 		req->page_descs[req->num_pages].length = this_num;
 		req->num_pages++;
 
author maintainer
#180 lore raw

PATCH 180/205 dmaengine: bcm2835: Fix interrupt race on RT

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Lukas Wunner <[email protected]>

commit f7da7782aba92593f7b82f03d2409a1c5f4db91b upstream.

If IRQ handlers are threaded (either because CONFIG_PREEMPT_RT_BASE is enabled or "threadirqs" was passed on the command line) and if system load is sufficiently high that wakeup latency of IRQ threads degrades, SPI DMA transactions on the BCM2835 occasionally break like this:

ks8851 spi0.0: SPI transfer timed out
bcm2835-dma 3f007000.dma: DMA transfer could not be terminated ks8851 spi0.0 eth2: ks8851_rdfifo: spi_sync() failed

The root cause is an assumption made by the DMA driver which is documented in a code comment in bcm2835_dma_terminate_all():

/*
* Stop DMA activity: we assume the callback will not be called
* after bcm_dma_abort() returns (even if it does, it will see
* c->desc is NULL and exit.)
*/

That assumption falls apart if the IRQ handler bcm2835_dma_callback() is
threaded: A client may terminate a descriptor and issue a new one before the IRQ handler had a chance to run. In fact the IRQ handler may miss an *arbitrary* number of descriptors. The result is the following race condition:

1. A descriptor finishes, its interrupt is deferred to the IRQ thread.
2. A client calls dma_terminate_async() which sets channel->desc = NULL.
3. The client issues a new descriptor. Because channel->desc is NULL, bcm2835_dma_issue_pending() immediately starts the descriptor.
4. Finally the IRQ thread runs and writes BCM2835_DMA_INT to the CS register to acknowledge the interrupt. This clears the ACTIVE flag, so the newly issued descriptor is paused in the middle of the transaction. Because channel->desc is not NULL, the IRQ thread finalizes the descriptor and tries to start the next one.

I see two possible solutions: The first is to call synchronize_irq() in bcm2835_dma_issue_pending() to wait until the IRQ thread has finished before issuing a new descriptor. The downside of this approach is unnecessary latency if clients desire rapidly terminating and re-issuing descriptors and don't have any use for an IRQ callback. (The SPI TX DMA channel is a case in point.)

A better alternative is to make the IRQ thread recognize that it has missed descriptors and avoid finalizing the newly issued descriptor. So first of all, set the ACTIVE flag when acknowledging the interrupt. This keeps a newly issued descriptor running.

If the descriptor was finished, the channel remains idle despite the ACTIVE flag being set. However the ACTIVE flag can then no longer be used to check whether the channel is idle, so instead check whether the register containing the current control block address is zero and finalize the current descriptor only if so.

That way, there is no impact on latency and throughput if the client doesn't care for the interrupt: Only minimal additional overhead is introduced for non-cyclic descriptors as one further MMIO read is necessary per interrupt to check for idleness of the channel. Cyclic descriptors are sped up slightly by removing one MMIO write per interrupt.

Fixes 96286b576690 ("dmaengine: Add support for BCM2835")
Cc [email protected] # v3.14+
Cc Frank Pavlic <[email protected]>
Cc Martin Sperl <[email protected]>
Cc Florian Meier <[email protected]>
Cc Clive Messer <[email protected]>
Cc Matthias Reichl <[email protected]>
---
 drivers/dma/bcm2835-dma.c |   33 ++++++++++++++++++---------------
 1 file changed, 18 insertions(+), 15 deletions(-)
drivers/dma/bcm2835-dma.c +18 -15
@@ -421,7 +421,12 @@ static int bcm2835_dma_abort(void __iome
 	long int timeout = 10000;
 
 	cs = readl(chan_base + BCM2835_DMA_CS);
-	if (!(cs & BCM2835_DMA_ACTIVE))
+
+	/*
+	 * A zero control block address means the channel is idle.
+	 * (The ACTIVE flag in the CS register is not a reliable indicator.)
+	 */
+	if (!readl(chan_base + BCM2835_DMA_ADDR))
 		return 0;
 
 	/* Write 0 to the active bit - Pause the DMA */
@@ -485,8 +490,15 @@ static irqreturn_t bcm2835_dma_callback(
 
 	spin_lock_irqsave(&c->vc.lock, flags);
 
-	/* Acknowledge interrupt */
-	writel(BCM2835_DMA_INT, c->chan_base + BCM2835_DMA_CS);
+	/*
+	 * Clear the INT flag to receive further interrupts. Keep the channel
+	 * active in case the descriptor is cyclic or in case the client has
+	 * already terminated the descriptor and issued a new one. (May happen
+	 * if this IRQ handler is threaded.) If the channel is finished, it
+	 * will remain idle despite the ACTIVE flag being set.
+	 */
+	writel(BCM2835_DMA_INT | BCM2835_DMA_ACTIVE,
+	       c->chan_base + BCM2835_DMA_CS);
 
 	d = c->desc;
 
@@ -494,11 +506,7 @@ static irqreturn_t bcm2835_dma_callback(
 		if (d->cyclic) {
 			/* call the cyclic callback */
 			vchan_cyclic_callback(&d->vd);
-
-			/* Keep the DMA engine running */
-			writel(BCM2835_DMA_ACTIVE,
-			       c->chan_base + BCM2835_DMA_CS);
-		} else {
+		} else if (!readl(c->chan_base + BCM2835_DMA_ADDR)) {
 			vchan_cookie_complete(&c->desc->vd);
 			bcm2835_dma_start_desc(c);
 		}
@@ -806,11 +814,7 @@ static int bcm2835_dma_terminate_all(str
 	list_del_init(&c->node);
 	spin_unlock(&d->lock);
 
-	/*
-	 * Stop DMA activity: we assume the callback will not be called
-	 * after bcm_dma_abort() returns (even if it does, it will see
-	 * c->desc is NULL and exit.)
-	 */
+	/* stop DMA activity */
 	if (c->desc) {
 		bcm2835_dma_desc_free(&c->desc->vd);
 		c->desc = NULL;
@@ -818,8 +822,7 @@ static int bcm2835_dma_terminate_all(str
 
 		/* Wait for stopping */
 		while (--timeout) {
-			if (!(readl(c->chan_base + BCM2835_DMA_CS) &
-						BCM2835_DMA_ACTIVE))
+			if (!readl(c->chan_base + BCM2835_DMA_ADDR))
 				break;
 
 			cpu_relax();
author maintainer
#181 lore raw

PATCH 181/205 dmaengine: bcm2835: Fix abort of transactions

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Lukas Wunner <[email protected]>

commit 9e528c799d17a4ac37d788c81440b50377dd592d upstream.

There are multiple issues with bcm2835_dma_abort() (which is called on termination of a transaction):

* The algorithm to abort the transaction first pauses the channel by clearing the ACTIVE flag in the CS register, then waits for the PAUSED flag to clear. Page 49 of the spec documents the latter as follows:

"Indicates if the DMA is currently paused and not transferring data.
This will occur if the active bit has been cleared [...]"
https://www.raspberrypi.org/app/uploads/2012/02/BCM2835-ARM-Peripherals.pdf

So the function is entering an infinite loop because it is waiting for PAUSED to clear which is always set due to the function having cleared the ACTIVE flag. The only thing that's saving it from itself is the upper bound of 10000 loop iterations.

The code comment says that the intention is to "wait for any current AXI transfer to complete", so the author probably wanted to check the WAITING_FOR_OUTSTANDING_WRITES flag instead. Amend the function accordingly.

* The CS register is only read at the beginning of the function. It needs to be read again after pausing the channel and before checking for outstanding writes, otherwise writes which were issued between the register read at the beginning of the function and pausing the channel may not be waited for.

* The function seeks to abort the transfer by writing 0 to the NEXTCONBK register and setting the ABORT and ACTIVE flags. Thereby, the 0 in NEXTCONBK is sought to be loaded into the CONBLK_AD register. However experimentation has shown this approach to not work: The CONBLK_AD register remains the same as before and the CS register contains 0x00000030 (PAUSED | DREQ_STOPS_DMA). In other words, the control block is not aborted but merely paused and it will be resumed once the next DMA transaction is started. That is absolutely not the desired behavior.

A simpler approach is to set the channel's RESET flag instead. This reliably zeroes the NEXTCONBK as well as the CS register. It requires less code and only a single MMIO write. This is also what popular user space DMA drivers do, e.g.:
https://github.com/metachris/RPIO/blob/master/source/c_pwm/pwm.c

Note that the spec is contradictory whether the NEXTCONBK register is writeable at all. On the one hand, page 41 claims:

"The value loaded into the NEXTCONBK register can be overwritten so that the linked list of Control Block data structures can be dynamically altered. However it is only safe to do this when the DMA is paused."

On the other hand, page 40 specifies:

"Only three registers in each channel's register set are directly writeable (CS, CONBLK_AD and DEBUG). The other registers (TI, SOURCE_AD, DEST_AD, TXFR_LEN, STRIDE & NEXTCONBK), are automatically loaded from a Control Block data structure held in external memory."

Fixes 96286b576690 ("dmaengine: Add support for BCM2835")
Cc [email protected] # v3.14+
Cc Frank Pavlic <[email protected]>
Cc Martin Sperl <[email protected]>
Cc Florian Meier <[email protected]>
Cc Clive Messer <[email protected]>
Cc Matthias Reichl <[email protected]>
---
 drivers/dma/bcm2835-dma.c |   41 +++++++++--------------------------------
 1 file changed, 9 insertions(+), 32 deletions(-)
drivers/dma/bcm2835-dma.c +9 -32
@@ -415,13 +415,11 @@ static void bcm2835_dma_fill_cb_chain_wi
 	}
 }
 
-static int bcm2835_dma_abort(void __iomem *chan_base)
+static int bcm2835_dma_abort(struct bcm2835_chan *c)
 {
-	unsigned long cs;
+	void __iomem *chan_base = c->chan_base;
 	long int timeout = 10000;
 
-	cs = readl(chan_base + BCM2835_DMA_CS);
-
 	/*
 	 * A zero control block address means the channel is idle.
 	 * (The ACTIVE flag in the CS register is not a reliable indicator.)
@@ -433,25 +431,16 @@ static int bcm2835_dma_abort(void __iome
 	writel(0, chan_base + BCM2835_DMA_CS);
 
 	/* Wait for any current AXI transfer to complete */
-	while ((cs & BCM2835_DMA_ISPAUSED) && --timeout) {
+	while ((readl(chan_base + BCM2835_DMA_CS) &
+		BCM2835_DMA_WAITING_FOR_WRITES) && --timeout)
 		cpu_relax();
-		cs = readl(chan_base + BCM2835_DMA_CS);
-	}
 
-	/* We'll un-pause when we set of our next DMA */
+	/* Peripheral might be stuck and fail to signal AXI write responses */
 	if (!timeout)
-		return -ETIMEDOUT;
-
-	if (!(cs & BCM2835_DMA_ACTIVE))
-		return 0;
-
-	/* Terminate the control block chain */
-	writel(0, chan_base + BCM2835_DMA_NEXTCB);
-
-	/* Abort the whole DMA */
-	writel(BCM2835_DMA_ABORT | BCM2835_DMA_ACTIVE,
-	       chan_base + BCM2835_DMA_CS);
+		dev_err(c->vc.chan.device->dev,
+			"failed to complete outstanding writes\n");
 
+	writel(BCM2835_DMA_RESET, chan_base + BCM2835_DMA_CS);
 	return 0;
 }
 
@@ -804,7 +793,6 @@ static int bcm2835_dma_terminate_all(str
 	struct bcm2835_chan *c = to_bcm2835_dma_chan(chan);
 	struct bcm2835_dmadev *d = to_bcm2835_dma_dev(c->vc.chan.device);
 	unsigned long flags;
-	int timeout = 10000;
 	LIST_HEAD(head);
 
 	spin_lock_irqsave(&c->vc.lock, flags);
@@ -818,18 +806,7 @@ static int bcm2835_dma_terminate_all(str
 	if (c->desc) {
 		bcm2835_dma_desc_free(&c->desc->vd);
 		c->desc = NULL;
-		bcm2835_dma_abort(c->chan_base);
-
-		/* Wait for stopping */
-		while (--timeout) {
-			if (!readl(c->chan_base + BCM2835_DMA_ADDR))
-				break;
-
-			cpu_relax();
-		}
-
-		if (!timeout)
-			dev_err(d->ddev.dev, "DMA transfer could not be terminated\n");
+		bcm2835_dma_abort(c);
 	}
 
 	vchan_get_all_descriptors(&c->vc, &head);
author maintainer
#182 lore raw

PATCH 182/205 dmaengine: imx-dma: fix wrong callback invoke

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Leonid Iziumtsev <[email protected]>

commit 341198eda723c8c1cddbb006a89ad9e362502ea2 upstream.

Once the "ld_queue" list is not empty, next descriptor will migrate into "ld_active" list. The "desc" variable will be overwritten during that transition. And later the dmaengine_desc_get_callback_invoke() will use it as an argument. As result we invoke wrong callback.

That behaviour was in place since:
commit fcaaba6c7136 ("dmaengine: imx-dma: fix callback path in tasklet"). But after commit 4cd13c21b207 ("softirq: Let ksoftirqd do its job") things got worse, since possible delay between tasklet_schedule() from DMA irq handler and actual tasklet function execution got bigger. And that gave more time for new DMA request to be submitted and to be put into "ld_queue" list.

It has been noticed that DMA issue is causing problems for "mxc-mmc" driver. While stressing the system with heavy network traffic and writing/reading to/from sd card simultaneously the timeout may happen:

10013000.sdhci: mxcmci_watchdog: read time out (status = 0x30004900)

That often lead to file system corruption.

---
 drivers/dma/imx-dma.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/dma/imx-dma.c +4 -4
@@ -623,7 +623,7 @@ static void imxdma_tasklet(unsigned long
 {
 	struct imxdma_channel *imxdmac = (void *)data;
 	struct imxdma_engine *imxdma = imxdmac->imxdma;
-	struct imxdma_desc *desc;
+	struct imxdma_desc *desc, *next_desc;
 	unsigned long flags;
 
 	spin_lock_irqsave(&imxdma->lock, flags);
@@ -653,10 +653,10 @@ static void imxdma_tasklet(unsigned long
 	list_move_tail(imxdmac->ld_active.next, &imxdmac->ld_free);
 
 	if (!list_empty(&imxdmac->ld_queue)) {
-		desc = list_first_entry(&imxdmac->ld_queue, struct imxdma_desc,
-					node);
+		next_desc = list_first_entry(&imxdmac->ld_queue,
+					     struct imxdma_desc, node);
 		list_move_tail(imxdmac->ld_queue.next, &imxdmac->ld_active);
-		if (imxdma_xfer_desc(desc) < 0)
+		if (imxdma_xfer_desc(next_desc) < 0)
 			dev_warn(imxdma->dev, "%s: channel: %d couldn't xfer desc\n",
 				 __func__, imxdmac->channel);
 	}
author maintainer
#183 lore raw

PATCH 183/205 futex: Handle early deadlock return correctly

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thomas Gleixner <[email protected]>

commit 1a1fb985f2e2b85ec0d3dc2e519ee48389ec2434 upstream.

commit 56222b212e8e ("futex: Drop hb->lock before enqueueing on the rtmutex") changed the locking rules in the futex code so that the hash bucket lock is not longer held while the waiter is enqueued into the rtmutex wait list. This made the lock and the unlock path symmetric, but unfortunately the possible early exit from __rt_mutex_proxy_start() due to a detected deadlock was not updated accordingly. That allows a concurrent unlocker to observe inconsitent state which triggers the warning in the unlock path.

futex_lock_pi() futex_unlock_pi()
lock(hb->lock)
queue(hb_waiter) lock(hb->lock)
lock(rtmutex->wait_lock)
unlock(hb->lock)

                                        // acquired hb->lock
                                        hb_waiter = futex_top_waiter()
                                        lock(rtmutex->wait_lock)
  __rt_mutex_proxy_start()
     ---> fail
          remove(rtmutex_waiter);
     ---> returns -EDEADLOCK
  unlock(rtmutex->wait_lock)
                                        // acquired wait_lock
                                        wake_futex_pi()
                                        rt_mutex_next_owner()
					  --> returns NULL
                                          --> WARN

lock(hb->lock)
unqueue(hb_waiter)

The problem is caused by the remove(rtmutex_waiter) in the failure case of __rt_mutex_proxy_start() as this lets the unlocker observe a waiter in the hash bucket but no waiter on the rtmutex, i.e. inconsistent state.

The original commit handles this correctly for the other early return cases (timeout, signal) by delaying the removal of the rtmutex waiter until the returning task reacquired the hash bucket lock.

Treat the failure case of __rt_mutex_proxy_start() in the same way and let the existing cleanup code handle the eventual handover of the rtmutex gracefully. The regular rt_mutex_proxy_start() gains the rtmutex waiter removal for the failure case, so that the other callsites are still operating correctly.

Add proper comments to the code so all these details are fully documented.

Thanks to Peter for helping with the analysis and writing the really valuable code comments.

---
 kernel/futex.c           |   28 ++++++++++++++++++----------
 kernel/locking/rtmutex.c |   37 ++++++++++++++++++++++++++++++++-----
 2 files changed, 50 insertions(+), 15 deletions(-)
kernel/futex.c +18 -10
@@ -2811,35 +2811,39 @@ retry_private:
 	 * and BUG when futex_unlock_pi() interleaves with this.
 	 *
 	 * Therefore acquire wait_lock while holding hb->lock, but drop the
-	 * latter before calling rt_mutex_start_proxy_lock(). This still fully
-	 * serializes against futex_unlock_pi() as that does the exact same
-	 * lock handoff sequence.
+	 * latter before calling __rt_mutex_start_proxy_lock(). This
+	 * interleaves with futex_unlock_pi() -- which does a similar lock
+	 * handoff -- such that the latter can observe the futex_q::pi_state
+	 * before __rt_mutex_start_proxy_lock() is done.
 	 */
 	raw_spin_lock_irq(&q.pi_state->pi_mutex.wait_lock);
 	spin_unlock(q.lock_ptr);
+	/*
+	 * __rt_mutex_start_proxy_lock() unconditionally enqueues the @rt_waiter
+	 * such that futex_unlock_pi() is guaranteed to observe the waiter when
+	 * it sees the futex_q::pi_state.
+	 */
 	ret = __rt_mutex_start_proxy_lock(&q.pi_state->pi_mutex, &rt_waiter, current);
 	raw_spin_unlock_irq(&q.pi_state->pi_mutex.wait_lock);
 
 	if (ret) {
 		if (ret == 1)
 			ret = 0;
-
-		spin_lock(q.lock_ptr);
-		goto no_block;
+		goto cleanup;
 	}
 
-
 	if (unlikely(to))
 		hrtimer_start_expires(&to->timer, HRTIMER_MODE_ABS);
 
 	ret = rt_mutex_wait_proxy_lock(&q.pi_state->pi_mutex, to, &rt_waiter);
 
+cleanup:
 	spin_lock(q.lock_ptr);
 	/*
-	 * If we failed to acquire the lock (signal/timeout), we must
+	 * If we failed to acquire the lock (deadlock/signal/timeout), we must
 	 * first acquire the hb->lock before removing the lock from the
-	 * rt_mutex waitqueue, such that we can keep the hb and rt_mutex
-	 * wait lists consistent.
+	 * rt_mutex waitqueue, such that we can keep the hb and rt_mutex wait
+	 * lists consistent.
 	 *
 	 * In particular; it is important that futex_unlock_pi() can not
 	 * observe this inconsistency.
@@ -2963,6 +2967,10 @@ retry:
 		 * there is no point where we hold neither; and therefore
 		 * wake_futex_pi() must observe a state consistent with what we
 		 * observed.
+		 *
+		 * In particular; this forces __rt_mutex_start_proxy() to
+		 * complete such that we're guaranteed to observe the
+		 * rt_waiter. Also see the WARN in wake_futex_pi().
 		 */
 		raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock);
 		spin_unlock(&hb->lock);
kernel/locking/rtmutex.c +32 -5
@@ -1726,12 +1726,33 @@ void rt_mutex_proxy_unlock(struct rt_mut
 	rt_mutex_set_owner(lock, NULL);
 }
 
+/**
+ * __rt_mutex_start_proxy_lock() - Start lock acquisition for another task
+ * @lock:		the rt_mutex to take
+ * @waiter:		the pre-initialized rt_mutex_waiter
+ * @task:		the task to prepare
+ *
+ * Starts the rt_mutex acquire; it enqueues the @waiter and does deadlock
+ * detection. It does not wait, see rt_mutex_wait_proxy_lock() for that.
+ *
+ * NOTE: does _NOT_ remove the @waiter on failure; must either call
+ * rt_mutex_wait_proxy_lock() or rt_mutex_cleanup_proxy_lock() after this.
+ *
+ * Returns:
+ *  0 - task blocked on lock
+ *  1 - acquired the lock for task, caller should wake it up
+ * <0 - error
+ *
+ * Special API call for PI-futex support.
+ */
 int __rt_mutex_start_proxy_lock(struct rt_mutex *lock,
 			      struct rt_mutex_waiter *waiter,
 			      struct task_struct *task)
 {
 	int ret;
 
+	lockdep_assert_held(&lock->wait_lock);
+
 	if (try_to_take_rt_mutex(lock, task, NULL))
 		return 1;
 
@@ -1749,9 +1770,6 @@ int __rt_mutex_start_proxy_lock(struct r
 		ret = 0;
 	}
 
-	if (unlikely(ret))
-		remove_waiter(lock, waiter);
-
 	debug_rt_mutex_print_deadlock(waiter);
 
 	return ret;
@@ -1763,12 +1781,18 @@ int __rt_mutex_start_proxy_lock(struct r
  * @waiter:		the pre-initialized rt_mutex_waiter
  * @task:		the task to prepare
  *
+ * Starts the rt_mutex acquire; it enqueues the @waiter and does deadlock
+ * detection. It does not wait, see rt_mutex_wait_proxy_lock() for that.
+ *
+ * NOTE: unlike __rt_mutex_start_proxy_lock this _DOES_ remove the @waiter
+ * on failure.
+ *
  * Returns:
  *  0 - task blocked on lock
  *  1 - acquired the lock for task, caller should wake it up
  * <0 - error
  *
- * Special API call for FUTEX_REQUEUE_PI support.
+ * Special API call for PI-futex support.
  */
 int rt_mutex_start_proxy_lock(struct rt_mutex *lock,
 			      struct rt_mutex_waiter *waiter,
@@ -1778,6 +1802,8 @@ int rt_mutex_start_proxy_lock(struct rt_
 
 	raw_spin_lock_irq(&lock->wait_lock);
 	ret = __rt_mutex_start_proxy_lock(lock, waiter, task);
+	if (unlikely(ret))
+		remove_waiter(lock, waiter);
 	raw_spin_unlock_irq(&lock->wait_lock);
 
 	return ret;
@@ -1845,7 +1871,8 @@ int rt_mutex_wait_proxy_lock(struct rt_m
  * @lock:		the rt_mutex we were woken on
  * @waiter:		the pre-initialized rt_mutex_waiter
  *
- * Attempt to clean up after a failed rt_mutex_wait_proxy_lock().
+ * Attempt to clean up after a failed __rt_mutex_start_proxy_lock() or
+ * rt_mutex_wait_proxy_lock().
  *
  * Unless we acquired the lock; we're still enqueued on the wait-list and can
  * in fact still be granted ownership until we're removed. Therefore we can
author maintainer
#184 lore raw

PATCH 184/205 irqchip/gic-v3-its: Plug allocation race for devices sharing a DevID

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marc Zyngier <[email protected]>

commit 9791ec7df0e7b4d80706ccea8f24b6542f6059e9 upstream.

On systems or VMs where multiple devices share a single DevID (because they sit behind a PCI bridge, or because the HW is broken in funky ways), we reuse the save its_device structure in order to reflect this.

It turns out that there is a distinct lack of locking when looking up the its_device, and two device being probed concurrently can result in double allocations. That's obviously not nice.

A solution for this is to have a per-ITS mutex that serializes device allocation.

A similar issue exists on the freeing side, which can run concurrently with the allocation. On top of now taking the appropriate lock, we also make sure that a shared device is never freed, as we have no way to currently track the life cycle of such object.

---
 drivers/irqchip/irq-gic-v3-its.c |   32 +++++++++++++++++++++++++++-----
 1 file changed, 27 insertions(+), 5 deletions(-)
drivers/irqchip/irq-gic-v3-its.c +27 -5
@@ -87,9 +87,14 @@ struct its_baser {
  * The ITS structure - contains most of the infrastructure, with the
  * top-level MSI domain, the command queue, the collections, and the
  * list of devices writing to it.
+ *
+ * dev_alloc_lock has to be taken for device allocations, while the
+ * spinlock must be taken to parse data structures such as the device
+ * list.
  */
 struct its_node {
 	raw_spinlock_t		lock;
+	struct mutex		dev_alloc_lock;
 	struct list_head	entry;
 	void __iomem		*base;
 	phys_addr_t		phys_base;
@@ -138,6 +143,7 @@ struct its_device {
 	void			*itt;
 	u32			nr_ites;
 	u32			device_id;
+	bool			shared;
 };
 
 static struct {
@@ -2109,6 +2115,7 @@ static int its_msi_prepare(struct irq_do
 	struct its_device *its_dev;
 	struct msi_domain_info *msi_info;
 	u32 dev_id;
+	int err = 0;
 
 	/*
 	 * We ignore "dev" entierely, and rely on the dev_id that has
@@ -2131,6 +2138,7 @@ static int its_msi_prepare(struct irq_do
 		return -EINVAL;
 	}
 
+	mutex_lock(&its->dev_alloc_lock);
 	its_dev = its_find_device(its, dev_id);
 	if (its_dev) {
 		/*
@@ -2138,18 +2146,22 @@ static int its_msi_prepare(struct irq_do
 		 * another alias (PCI bridge of some sort). No need to
 		 * create the device.
 		 */
+		its_dev->shared = true;
 		pr_debug("Reusing ITT for devID %x\n", dev_id);
 		goto out;
 	}
 
 	its_dev = its_create_device(its, dev_id, nvec, true);
-	if (!its_dev)
-		return -ENOMEM;
+	if (!its_dev) {
+		err = -ENOMEM;
+		goto out;
+	}
 
 	pr_debug("ITT %d entries, %d bits\n", nvec, ilog2(nvec));
 out:
+	mutex_unlock(&its->dev_alloc_lock);
 	info->scratchpad[0].ptr = its_dev;
-	return 0;
+	return err;
 }
 
 static struct msi_domain_ops its_msi_domain_ops = {
@@ -2252,6 +2264,7 @@ static void its_irq_domain_free(struct i
 {
 	struct irq_data *d = irq_domain_get_irq_data(domain, virq);
 	struct its_device *its_dev = irq_data_get_irq_chip_data(d);
+	struct its_node *its = its_dev->its;
 	int i;
 
 	for (i = 0; i < nr_irqs; i++) {
@@ -2266,8 +2279,14 @@ static void its_irq_domain_free(struct i
 		irq_domain_reset_irq_data(data);
 	}
 
-	/* If all interrupts have been freed, start mopping the floor */
-	if (bitmap_empty(its_dev->event_map.lpi_map,
+	mutex_lock(&its->dev_alloc_lock);
+
+	/*
+	 * If all interrupts have been freed, start mopping the
+	 * floor. This is conditionned on the device not being shared.
+	 */
+	if (!its_dev->shared &&
+	    bitmap_empty(its_dev->event_map.lpi_map,
 			 its_dev->event_map.nr_lpis)) {
 		its_lpi_free_chunks(its_dev->event_map.lpi_map,
 				    its_dev->event_map.lpi_base,
@@ -2279,6 +2298,8 @@ static void its_irq_domain_free(struct i
 		its_free_device(its_dev);
 	}
 
+	mutex_unlock(&its->dev_alloc_lock);
+
 	irq_domain_free_irqs_parent(domain, virq, nr_irqs);
 }
 
@@ -2966,6 +2987,7 @@ static int __init its_probe_one(struct r
 	}
 
 	raw_spin_lock_init(&its->lock);
+	mutex_init(&its->dev_alloc_lock);
 	INIT_LIST_HEAD(&its->entry);
 	INIT_LIST_HEAD(&its->its_device_list);
 	typer = gic_read_typer(its_base + GITS_TYPER);
author maintainer
#185 lore raw

PATCH 185/205 usb: phy: am335x: fix race condition in _probe

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bin Liu <[email protected]>

commit a53469a68eb886e84dd8b69a1458a623d3591793 upstream.

power off the phy should be done before populate the phy. Otherwise, am335x_init() could be called by the phy owner to power on the phy first, then am335x_phy_probe() turns off the phy again without the caller knowing it.

Fixes 2fc711d76352 ("usb: phy: am335x: Enable USB remote wakeup using PHY wakeup")
Cc [email protected] # v3.18+
---
 drivers/usb/phy/phy-am335x.c |    5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
drivers/usb/phy/phy-am335x.c +1 -4
@@ -60,9 +60,6 @@ static int am335x_phy_probe(struct platf
 	if (ret)
 		return ret;
 
-	ret = usb_add_phy_dev(&am_phy->usb_phy_gen.phy);
-	if (ret)
-		return ret;
 	am_phy->usb_phy_gen.phy.init = am335x_init;
 	am_phy->usb_phy_gen.phy.shutdown = am335x_shutdown;
 
@@ -81,7 +78,7 @@ static int am335x_phy_probe(struct platf
 	device_set_wakeup_enable(dev, false);
 	phy_ctrl_power(am_phy->phy_ctrl, am_phy->id, am_phy->dr_mode, false);
 
-	return 0;
+	return usb_add_phy_dev(&am_phy->usb_phy_gen.phy);
 }
 
 static int am335x_phy_remove(struct platform_device *pdev)
author maintainer
#186 lore raw

PATCH 186/205 usb: dwc3: gadget: Handle 0 xfer length for OUT EP

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tejas Joglekar <[email protected]>

commit 1e19cdc8060227b0802bda6bc0bd22b23679ba32 upstream.

For OUT endpoints, zero-length transfers require MaxPacketSize buffer as per the DWC_usb3 programming guide 3.30a section 4.2.3.3.

This patch fixes this by explicitly checking zero length transfer to correctly pad up to MaxPacketSize.

Fixes c6267a51639b ("usb: dwc3: gadget: align transfers to wMaxPacketSize")
---
 drivers/usb/dwc3/gadget.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/dwc3/gadget.c +1 -1
@@ -1114,7 +1114,7 @@ static void dwc3_prepare_one_trb_linear(
 	unsigned int maxp = usb_endpoint_maxp(dep->endpoint.desc);
 	unsigned int rem = length % maxp;
 
-	if (rem && usb_endpoint_dir_out(dep->endpoint.desc)) {
+	if ((!length || rem) && usb_endpoint_dir_out(dep->endpoint.desc)) {
 		struct dwc3	*dwc = dep->dwc;
 		struct dwc3_trb	*trb;
 
author maintainer
#187 lore raw

PATCH 187/205 usb: gadget: udc: net2272: Fix bitwise and boolean operations

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Gustavo A. R. Silva <[email protected]>

commit 07c69f1148da7de3978686d3af9263325d9d60bd upstream.

(!x & y) strikes again.

Fix bitwise and boolean operations by enclosing the expression:

	intcsr & (1 << NET2272_PCI_IRQ)

in parentheses, before applying the boolean operator '!'.

Notice that this code has been there since 2011. So, it would be helpful if someone can double-check this.

This issue was detected with the help of Coccinelle.

Fixes ceb80363b2ec ("USB: net2272: driver for PLX NET2272 USB device controller")
---
 drivers/usb/gadget/udc/net2272.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/gadget/udc/net2272.c +1 -1
@@ -2096,7 +2096,7 @@ static irqreturn_t net2272_irq(int irq,
 #if defined(PLX_PCI_RDK2)
 	/* see if PCI int for us by checking irqstat */
 	intcsr = readl(dev->rdk2.fpga_base_addr + RDK2_IRQSTAT);
-	if (!intcsr & (1 << NET2272_PCI_IRQ)) {
+	if (!(intcsr & (1 << NET2272_PCI_IRQ))) {
 		spin_unlock(&dev->lock);
 		return IRQ_NONE;
 	}
author maintainer
#188 lore raw

PATCH 188/205 usb: gadget: musb: fix short isoc packets with inventra dma

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paul Elder <[email protected]>

commit c418fd6c01fbc5516a2cd1eaf1df1ec86869028a upstream.

Handling short packets (length < max packet size) in the Inventra DMA engine in the MUSB driver causes the MUSB DMA controller to hang. An example of a problem that is caused by this problem is when streaming video out of a UVC gadget, only the first video frame is transferred.

For short packets (mode-0 or mode-1 DMA), MUSB_TXCSR_TXPKTRDY must be set manually by the driver. This was previously done in musb_g_tx (musb_gadget.c), but incorrectly (all csr flags were cleared, and only MUSB_TXCSR_MODE and MUSB_TXCSR_TXPKTRDY were set). Fixing that problem allows some requests to be transferred correctly, but multiple requests were often put together in one USB packet, and caused problems if the packet size was not a multiple of 4. Instead, set MUSB_TXCSR_TXPKTRDY in dma_controller_irq (musbhsdma.c), just like host mode transfers.

This topic was originally tackled by Nicolas Boichat [0] [1] and is discussed further at [2] as part of his GSoC project [3].

[0] https://groups.google.com/forum/?hl=en#!topic/beagleboard-gsoc/k8Azwfp75CU
[1] https://gitorious.org/beagleboard-usbsniffer/beagleboard-usbsniffer-kernel/commit/b0be3b6cc195ba732189b04f1d43ec843c3e54c9?p=beagleboard-usbsniffer:beagleboard-usbsniffer-kernel.git;a=patch;h=b0be3b6cc195ba732189b04f1d43ec843c3e54c9
[2] http://beagleboard-usbsniffer.blogspot.com/2010/07/musb-isochronous-transfers-fixed.html
[3] http://elinux.org/BeagleBoard/GSoC/USBSniffer

Fixes 550a7375fe72 ("USB: Add MUSB and TUSB support")
Cc stable <[email protected]>
---
 drivers/usb/musb/musb_gadget.c |   13 +------------
 drivers/usb/musb/musbhsdma.c   |   21 +++++++++++----------
 2 files changed, 12 insertions(+), 22 deletions(-)
drivers/usb/musb/musb_gadget.c +1 -12
@@ -477,13 +477,10 @@ void musb_g_tx(struct musb *musb, u8 epn
 	}
 
 	if (request) {
-		u8	is_dma = 0;
-		bool	short_packet = false;
 
 		trace_musb_req_tx(req);
 
 		if (dma && (csr & MUSB_TXCSR_DMAENAB)) {
-			is_dma = 1;
 			csr |= MUSB_TXCSR_P_WZC_BITS;
 			csr &= ~(MUSB_TXCSR_DMAENAB | MUSB_TXCSR_P_UNDERRUN |
 				 MUSB_TXCSR_TXPKTRDY | MUSB_TXCSR_AUTOSET);
@@ -501,16 +498,8 @@ void musb_g_tx(struct musb *musb, u8 epn
 		 */
 		if ((request->zero && request->length)
 			&& (request->length % musb_ep->packet_sz == 0)
-			&& (request->actual == request->length))
-				short_packet = true;
+			&& (request->actual == request->length)) {
 
-		if ((musb_dma_inventra(musb) || musb_dma_ux500(musb)) &&
-			(is_dma && (!dma->desired_mode ||
-				(request->actual &
-					(musb_ep->packet_sz - 1)))))
-				short_packet = true;
-
-		if (short_packet) {
 			/*
 			 * On DMA completion, FIFO may not be
 			 * available yet...
drivers/usb/musb/musbhsdma.c +11 -10
@@ -320,12 +320,10 @@ static irqreturn_t dma_controller_irq(in
 				channel->status = MUSB_DMA_STATUS_FREE;
 
 				/* completed */
-				if ((devctl & MUSB_DEVCTL_HM)
-					&& (musb_channel->transmit)
-					&& ((channel->desired_mode == 0)
-					    || (channel->actual_len &
-					    (musb_channel->max_packet_sz - 1)))
-				    ) {
+				if (musb_channel->transmit &&
+					(!channel->desired_mode ||
+					(channel->actual_len %
+					    musb_channel->max_packet_sz))) {
 					u8  epnum  = musb_channel->epnum;
 					int offset = musb->io.ep_offset(epnum,
 								    MUSB_TXCSR);
@@ -337,11 +335,14 @@ static irqreturn_t dma_controller_irq(in
 					 */
 					musb_ep_select(mbase, epnum);
 					txcsr = musb_readw(mbase, offset);
-					txcsr &= ~(MUSB_TXCSR_DMAENAB
+					if (channel->desired_mode == 1) {
+						txcsr &= ~(MUSB_TXCSR_DMAENAB
 							| MUSB_TXCSR_AUTOSET);
-					musb_writew(mbase, offset, txcsr);
-					/* Send out the packet */
-					txcsr &= ~MUSB_TXCSR_DMAMODE;
+						musb_writew(mbase, offset, txcsr);
+						/* Send out the packet */
+						txcsr &= ~MUSB_TXCSR_DMAMODE;
+						txcsr |= MUSB_TXCSR_DMAENAB;
+					}
 					txcsr |=  MUSB_TXCSR_TXPKTRDY;
 					musb_writew(mbase, offset, txcsr);
 				}
author maintainer
#189 lore raw

PATCH 189/205 staging: speakup: fix tty-operation NULL derefs

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit a1960e0f1639cb1f7a3d94521760fc73091f6640 upstream.

The send_xchar() and tiocmset() tty operations are optional. Add the missing sanity checks to prevent user-space triggerable NULL-pointer dereferences.

Fixes 6b9ad1c742bf ("staging: speakup: add send_xchar, tiocmset and input functionality for tty")
Cc stable <[email protected]> # 4.13
Cc Okash Khawaja <[email protected]>
Cc Samuel Thibault <[email protected]>
---
 drivers/staging/speakup/spk_ttyio.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/staging/speakup/spk_ttyio.c +4 -2
@@ -246,7 +246,8 @@ static void spk_ttyio_send_xchar(char ch
 		return;
 	}
 
-	speakup_tty->ops->send_xchar(speakup_tty, ch);
+	if (speakup_tty->ops->send_xchar)
+		speakup_tty->ops->send_xchar(speakup_tty, ch);
 	mutex_unlock(&speakup_tty_mutex);
 }
 
@@ -258,7 +259,8 @@ static void spk_ttyio_tiocmset(unsigned
 		return;
 	}
 
-	speakup_tty->ops->tiocmset(speakup_tty, set, clear);
+	if (speakup_tty->ops->tiocmset)
+		speakup_tty->ops->tiocmset(speakup_tty, set, clear);
 	mutex_unlock(&speakup_tty_mutex);
 }
 
author maintainer
#190 lore raw

PATCH 190/205 scsi: cxlflash: Prevent deadlock when adapter probe fails

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vaibhav Jain <[email protected]>

commit bb61b843ffd46978d7ca5095453e572714934eeb upstream.

Presently when an error is encountered during probe of the cxlflash adapter, a deadlock is seen with cpu thread stuck inside cxlflash_remove(). Below is the trace of the deadlock as logged by khungtaskd:

cxlflash 0006:00:00.0: cxlflash_probe: init_afu failed rc=-16

INFO: task kworker/80:1:890 blocked for more than 120 seconds.
       Not tainted 5.0.0-rc4-capi2-kexec+ #2
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
kworker/80:1    D    0   890      2 0x00000808
Workqueue: events work_for_cpu_fn

Call Trace:
 0x4d72136320 (unreliable)
 __switch_to+0x2cc/0x460
 __schedule+0x2bc/0xac0
 schedule+0x40/0xb0
 cxlflash_remove+0xec/0x640 [cxlflash]
 cxlflash_probe+0x370/0x8f0 [cxlflash]
 local_pci_probe+0x6c/0x140
 work_for_cpu_fn+0x38/0x60
 process_one_work+0x260/0x530
 worker_thread+0x280/0x5d0
 kthread+0x1a8/0x1b0
 ret_from_kernel_thread+0x5c/0x80
INFO: task systemd-udevd:5160 blocked for more than 120 seconds.

The deadlock occurs as cxlflash_remove() is called from cxlflash_probe() without setting 'cxlflash_cfg->state' to STATE_PROBED and the probe thread starts to wait on 'cxlflash_cfg->reset_waitq'. Since the device was never successfully probed the 'cxlflash_cfg->state' never changes from STATE_PROBING hence the deadlock occurs.

We fix this deadlock by setting the variable 'cxlflash_cfg->state' to STATE_PROBED in case an error occurs during cxlflash_probe() and just before calling cxlflash_remove().

Fixes c21e0bbfc485("cxlflash: Base support for IBM CXL Flash Adapter")
---
 drivers/scsi/cxlflash/main.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/scsi/cxlflash/main.c +2 -0
@@ -3659,6 +3659,7 @@ static int cxlflash_probe(struct pci_dev
 	host->max_cmd_len = CXLFLASH_MAX_CDB_LEN;
 
 	cfg = shost_priv(host);
+	cfg->state = STATE_PROBING;
 	cfg->host = host;
 	rc = alloc_mem(cfg);
 	if (rc) {
@@ -3741,6 +3742,7 @@ out:
 	return rc;
 
 out_remove:
+	cfg->state = STATE_PROBED;
 	cxlflash_remove(pdev);
 	goto out;
 }
author maintainer
#191 lore raw

PATCH 191/205 scsi: aic94xx: fix module loading

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: James Bottomley <[email protected]>

commit 42caa0edabd6a0a392ec36a5f0943924e4954311 upstream.

The aic94xx driver is currently failing to load with errors like

sysfs: cannot create duplicate filename '/devices/pci0000:00/0000:00:03.0/0000:02:00.3/0000:07:02.0/revision'

Because the PCI code had recently added a file named 'revision' to every PCI device. Fix this by renaming the aic94xx revision file to aic_revision. This is safe to do for us because as far as I can tell, there's nothing in userspace relying on the current aic94xx revision file so it can be renamed without breaking anything.

Fixes 702ed3be1b1b (PCI: Create revision file in sysfs)
---
 drivers/scsi/aic94xx/aic94xx_init.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/scsi/aic94xx/aic94xx_init.c +4 -4
@@ -281,7 +281,7 @@ static ssize_t asd_show_dev_rev(struct d
 	return snprintf(buf, PAGE_SIZE, "%s\n",
 			asd_dev_rev[asd_ha->revision_id]);
 }
-static DEVICE_ATTR(revision, S_IRUGO, asd_show_dev_rev, NULL);
+static DEVICE_ATTR(aic_revision, S_IRUGO, asd_show_dev_rev, NULL);
 
 static ssize_t asd_show_dev_bios_build(struct device *dev,
 				       struct device_attribute *attr,char *buf)
@@ -478,7 +478,7 @@ static int asd_create_dev_attrs(struct a
 {
 	int err;
 
-	err = device_create_file(&asd_ha->pcidev->dev, &dev_attr_revision);
+	err = device_create_file(&asd_ha->pcidev->dev, &dev_attr_aic_revision);
 	if (err)
 		return err;
 
@@ -500,13 +500,13 @@ err_update_bios:
 err_biosb:
 	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_bios_build);
 err_rev:
-	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_revision);
+	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_aic_revision);
 	return err;
 }
 
 static void asd_remove_dev_attrs(struct asd_ha_struct *asd_ha)
 {
-	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_revision);
+	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_aic_revision);
 	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_bios_build);
 	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_pcba_sn);
 	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_update_bios);
author maintainer
#192 lore raw

PATCH 192/205 KVM: x86: work around leak of uninitialized stack contents (CVE-2019-7222)

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paolo Bonzini <[email protected]>

commit 353c0956a618a07ba4bbe7ad00ff29fe70e8412a upstream.

Bugzilla: 1671930

Emulation of certain instructions (VMXON, VMCLEAR, VMPTRLD, VMWRITE with memory operand, INVEPT, INVVPID) can incorrectly inject a page fault when passed an operand that points to an MMIO address. The page fault will use uninitialized kernel stack memory as the CR2 and error code.

The right behavior would be to abort the VM with a KVM_EXIT_INTERNAL_ERROR exit to userspace; however, it is not an easy fix, so for now just ensure that the error code and CR2 are zero.

Embargoed until Feb 7th 2019.

---
 arch/x86/kvm/x86.c |    7 +++++++
 1 file changed, 7 insertions(+)
arch/x86/kvm/x86.c +7 -0
@@ -4611,6 +4611,13 @@ int kvm_read_guest_virt(struct kvm_vcpu
 {
 	u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0;
 
+	/*
+	 * FIXME: this should call handle_emulation_failure if X86EMUL_IO_NEEDED
+	 * is returned, but our callers are not ready for that and they blindly
+	 * call kvm_inject_page_fault.  Ensure that they at least do not leak
+	 * uninitialized kernel stack memory into cr2 and error code.
+	 */
+	memset(exception, 0, sizeof(*exception));
 	return kvm_read_guest_virt_helper(addr, val, bytes, vcpu, access,
 					  exception);
 }
author maintainer
#193 lore raw

PATCH 193/205 kvm: fix kvm_ioctl_create_device() reference counting (CVE-2019-6974)

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jann Horn <[email protected]>

commit cfa39381173d5f969daf43582c95ad679189cbc9 upstream.

kvm_ioctl_create_device() does the following:

1. creates a device that holds a reference to the VM object (with a borrowed reference, the VM's refcount has not been bumped yet)
2. initializes the device
3. transfers the reference to the device to the caller's file descriptor table
4. calls kvm_get_kvm() to turn the borrowed reference to the VM into a real reference

The ownership transfer in step 3 must not happen before the reference to the VM becomes a proper, non-borrowed reference, which only happens in step 4. After step 3, an attacker can close the file descriptor and drop the borrowed reference, which can cause the refcount of the kvm object to drop to zero.

This means that we need to grab a reference for the device before anon_inode_getfd(), otherwise the VM can disappear from under us.

Fixes 852b6d57dc7f ("kvm: add device control API")
---
 virt/kvm/kvm_main.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
virt/kvm/kvm_main.c +2 -1
@@ -2912,8 +2912,10 @@ static int kvm_ioctl_create_device(struc
 	if (ops->init)
 		ops->init(dev);
 
+	kvm_get_kvm(kvm);
 	ret = anon_inode_getfd(ops->name, &kvm_device_fops, dev, O_RDWR | O_CLOEXEC);
 	if (ret < 0) {
+		kvm_put_kvm(kvm);
 		mutex_lock(&kvm->lock);
 		list_del(&dev->vm_node);
 		mutex_unlock(&kvm->lock);
@@ -2921,7 +2923,6 @@ static int kvm_ioctl_create_device(struc
 		return ret;
 	}
 
-	kvm_get_kvm(kvm);
 	cd->fd = ret;
 	return 0;
 }
author maintainer
#194 lore raw

PATCH 194/205 KVM: nVMX: unconditionally cancel preemption timer in free_nested (CVE-2019-7221)

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Peter Shier <[email protected]>

commit ecec76885bcfe3294685dc363fd1273df0d5d65f upstream.

Bugzilla: 1671904

There are multiple code paths where an hrtimer may have been started to emulate an L1 VMX preemption timer that can result in a call to free_nested without an intervening L2 exit where the hrtimer is normally cancelled. Unconditionally cancel in free_nested to cover all cases.

Embargoed until Feb 7th 2019.

Message-Id: <[email protected]>

---
 arch/x86/kvm/vmx.c |    1 +
 1 file changed, 1 insertion(+)
arch/x86/kvm/vmx.c +1 -0
@@ -7708,6 +7708,7 @@ static void free_nested(struct vcpu_vmx
 	if (!vmx->nested.vmxon)
 		return;
 
+	hrtimer_cancel(&vmx->nested.preemption_timer);
 	vmx->nested.vmxon = false;
 	free_vpid(vmx->nested.vpid02);
 	vmx->nested.posted_intr_nv = -1;
author maintainer
#195 lore raw

PATCH 195/205 cpu/hotplug: Fix "SMT disabled by BIOS" detection for KVM

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Josh Poimboeuf <[email protected]>

commit b284909abad48b07d3071a9fc9b5692b3e64914b upstream.

With the following commit:

73d5e2b47264 ("cpu/hotplug: detect SMT disabled by BIOS")

... the hotplug code attempted to detect when SMT was disabled by BIOS, in which case it reported SMT as permanently disabled. However, that code broke a virt hotplug scenario, where the guest is booted with only primary CPU threads, and a sibling is brought online later.

The problem is that there doesn't seem to be a way to reliably distinguish between the HW "SMT disabled by BIOS" case and the virt "sibling not yet brought online" case. So the above-mentioned commit was a bit misguided, as it permanently disabled SMT for both cases, preventing future virt sibling hotplugs.

Going back and reviewing the original problems which were attempted to be solved by that commit, when SMT was disabled in BIOS:

1) /sys/devices/system/cpu/smt/control showed "on" instead of "notsupported"; and

2) vmx_vm_init() was incorrectly showing the L1TF_MSG_SMT warning.

I'd propose that we instead consider #1 above to not actually be a problem. Because, at least in the virt case, it's possible that SMT wasn't disabled by BIOS and a sibling thread could be brought online later. So it makes sense to just always default the smt control to "on" to allow for that possibility (assuming cpuid indicates that the CPU supports SMT).

The real problem is #2, which has a simple fix: change vmx_vm_init() to query the actual current SMT state -- i.e., whether any siblings are currently online -- instead of looking at the SMT "control" sysfs value.

So fix it by:

a) reverting the original "fix" and its followup fix:

73d5e2b47264 ("cpu/hotplug: detect SMT disabled by BIOS") bc2d8d262cba ("cpu/hotplug: Fix SMT supported evaluation")

and

b) changing vmx_vm_init() to query the actual current SMT state -- instead of the sysfs control value -- to determine whether the L1TF warning is needed. This also requires the 'sched_smt_present' variable to exported, instead of 'cpu_smt_control'.

---
 arch/x86/kernel/cpu/bugs.c |    2 +-
 arch/x86/kvm/vmx.c         |    3 ++-
 include/linux/cpu.h        |    2 --
 kernel/cpu.c               |   33 ++++-----------------------------
 kernel/sched/fair.c        |    1 +
 kernel/smp.c               |    2 --
 6 files changed, 8 insertions(+), 35 deletions(-)
arch/x86/kernel/cpu/bugs.c +1 -1
@@ -68,7 +68,7 @@ void __init check_bugs(void)
 	 * identify_boot_cpu() initialized SMT support information, let the
 	 * core code know.
 	 */
-	cpu_smt_check_topology_early();
+	cpu_smt_check_topology();
 
 	if (!IS_ENABLED(CONFIG_SMP)) {
 		pr_info("CPU: ");
arch/x86/kvm/vmx.c +2 -1
@@ -27,6 +27,7 @@
 #include <linux/mm.h>
 #include <linux/highmem.h>
 #include <linux/sched.h>
+#include <linux/sched/smt.h>
 #include <linux/moduleparam.h>
 #include <linux/mod_devicetable.h>
 #include <linux/trace_events.h>
@@ -10120,7 +10121,7 @@ static int vmx_vm_init(struct kvm *kvm)
 			 * Warn upon starting the first VM in a potentially
 			 * insecure environment.
 			 */
-			if (cpu_smt_control == CPU_SMT_ENABLED)
+			if (sched_smt_active())
 				pr_warn_once(L1TF_MSG_SMT);
 			if (l1tf_vmx_mitigation == VMENTER_L1D_FLUSH_NEVER)
 				pr_warn_once(L1TF_MSG_L1D);
include/linux/cpu.h +0 -2
@@ -188,12 +188,10 @@ enum cpuhp_smt_control {
 #if defined(CONFIG_SMP) && defined(CONFIG_HOTPLUG_SMT)
 extern enum cpuhp_smt_control cpu_smt_control;
 extern void cpu_smt_disable(bool force);
-extern void cpu_smt_check_topology_early(void);
 extern void cpu_smt_check_topology(void);
 #else
 # define cpu_smt_control		(CPU_SMT_ENABLED)
 static inline void cpu_smt_disable(bool force) { }
-static inline void cpu_smt_check_topology_early(void) { }
 static inline void cpu_smt_check_topology(void) { }
 #endif
 
kernel/cpu.c +4 -29
@@ -356,9 +356,6 @@ void __weak arch_smt_update(void) { }
 
 #ifdef CONFIG_HOTPLUG_SMT
 enum cpuhp_smt_control cpu_smt_control __read_mostly = CPU_SMT_ENABLED;
-EXPORT_SYMBOL_GPL(cpu_smt_control);
-
-static bool cpu_smt_available __read_mostly;
 
 void __init cpu_smt_disable(bool force)
 {
@@ -376,25 +373,11 @@ void __init cpu_smt_disable(bool force)
 
 /*
  * The decision whether SMT is supported can only be done after the full
- * CPU identification. Called from architecture code before non boot CPUs
- * are brought up.
- */
-void __init cpu_smt_check_topology_early(void)
-{
-	if (!topology_smt_supported())
-		cpu_smt_control = CPU_SMT_NOT_SUPPORTED;
-}
-
-/*
- * If SMT was disabled by BIOS, detect it here, after the CPUs have been
- * brought online. This ensures the smt/l1tf sysfs entries are consistent
- * with reality. cpu_smt_available is set to true during the bringup of non
- * boot CPUs when a SMT sibling is detected. Note, this may overwrite
- * cpu_smt_control's previous setting.
+ * CPU identification. Called from architecture code.
  */
 void __init cpu_smt_check_topology(void)
 {
-	if (!cpu_smt_available)
+	if (!topology_smt_supported())
 		cpu_smt_control = CPU_SMT_NOT_SUPPORTED;
 }
 
@@ -407,18 +390,10 @@ early_param("nosmt", smt_cmdline_disable
 
 static inline bool cpu_smt_allowed(unsigned int cpu)
 {
-	if (topology_is_primary_thread(cpu))
+	if (cpu_smt_control == CPU_SMT_ENABLED)
 		return true;
 
-	/*
-	 * If the CPU is not a 'primary' thread and the booted_once bit is
-	 * set then the processor has SMT support. Store this information
-	 * for the late check of SMT support in cpu_smt_check_topology().
-	 */
-	if (per_cpu(cpuhp_state, cpu).booted_once)
-		cpu_smt_available = true;
-
-	if (cpu_smt_control == CPU_SMT_ENABLED)
+	if (topology_is_primary_thread(cpu))
 		return true;
 
 	/*
kernel/sched/fair.c +1 -0
@@ -5651,6 +5651,7 @@ find_idlest_cpu(struct sched_group *grou
 
 #ifdef CONFIG_SCHED_SMT
 DEFINE_STATIC_KEY_FALSE(sched_smt_present);
+EXPORT_SYMBOL_GPL(sched_smt_present);
 
 static inline void set_idle_cores(int cpu, int val)
 {
kernel/smp.c +0 -2
@@ -584,8 +584,6 @@ void __init smp_init(void)
 		num_nodes, (num_nodes > 1 ? "s" : ""),
 		num_cpus,  (num_cpus  > 1 ? "s" : ""));
 
-	/* Final decision about SMT support */
-	cpu_smt_check_topology();
 	/* Any cleanup work */
 	smp_cpus_done(setup_max_cpus);
 }
author maintainer
#196 lore raw

PATCH 196/205 perf/x86/intel/uncore: Add Node ID mask

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kan Liang <[email protected]>

commit 9e63a7894fd302082cf3627fe90844421a6cbe7f upstream.

Some PCI uncore PMUs cannot be registered on an 8-socket system (HPE Superdome Flex).

To understand which Socket the PCI uncore PMUs belongs to, perf retrieves the local Node ID of the uncore device from CPUNODEID(0xC0) of the PCI configuration space, and the mapping between Socket ID and Node ID from GIDNIDMAP(0xD4). The Socket ID can be calculated accordingly.

The local Node ID is only available at bit 2:0, but current code doesn't mask it. If a BIOS doesn't clear the rest of the bits, an incorrect Node ID will be fetched.

Filter the Node ID by adding a mask.

Cc Alexander Shishkin <[email protected]>
Cc Arnaldo Carvalho de Melo <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc <[email protected]> # v3.7+
Fixes 7c94ee2e0917 ("perf/x86: Add Intel Nehalem and Sandy Bridge-EP uncore support")
---
 arch/x86/events/intel/uncore_snbep.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
arch/x86/events/intel/uncore_snbep.c +3 -1
@@ -1221,6 +1221,8 @@ static struct pci_driver snbep_uncore_pc
 	.id_table	= snbep_uncore_pci_ids,
 };
 
+#define NODE_ID_MASK	0x7
+
 /*
  * build pci bus to socket mapping
  */
@@ -1242,7 +1244,7 @@ static int snbep_pci2phy_map_init(int de
 		err = pci_read_config_dword(ubox_dev, nodeid_loc, &config);
 		if (err)
 			break;
-		nodeid = config;
+		nodeid = config & NODE_ID_MASK;
 		/* get the Node ID mapping */
 		err = pci_read_config_dword(ubox_dev, idmap_loc, &config);
 		if (err)
author maintainer
#197 lore raw

PATCH 197/205 x86/MCE: Initialize mce.bank in the case of a fatal error in mce_no_way_out()

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tony Luck <[email protected]>

commit d28af26faa0b1daf3c692603d46bc4687c16f19e upstream.

Internal injection testing crashed with a console log that said:

mce: [Hardware Error]: CPU 7: Machine Check Exception: f Bank 0: bd80000000100134

This caused a lot of head scratching because the MCACOD (bits 15:0) of that status is a signature from an L1 data cache error. But Linux says that it found it in "Bank 0", which on this model CPU only reports L1 instruction cache errors.

The answer was that Linux doesn't initialize "m->bank" in the case that it finds a fatal error in the mce_no_way_out() pre-scan of banks. If this was a local machine check, then this partially initialized struct mce is being passed to mce_panic().

Fix is simple: just initialize m->bank in the case of a fatal error.

Fixes 40c36e2741d7 ("x86/mce: Fix incorrect "Machine check from unknown source" message")
Cc "H. Peter Anvin" <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Vishal Verma <[email protected]>
Cc x86-ml <[email protected]>
Cc [email protected] # v4.18 Note pre-v5.0 arch/x86/kernel/cpu/mce/core.c was called arch/x86/kernel/cpu/mcheck/mce.c
---
 arch/x86/kernel/cpu/mcheck/mce.c |    1 +
 1 file changed, 1 insertion(+)
arch/x86/kernel/cpu/mcheck/mce.c +1 -0
@@ -773,6 +773,7 @@ static int mce_no_way_out(struct mce *m,
 			quirk_no_way_out(i, m, regs);
 
 		if (mce_severity(m, mca_cfg.tolerant, &tmp, true) >= MCE_PANIC_SEVERITY) {
+			m->bank = i;
 			mce_read_aux(m, i);
 			*msg = tmp;
 			return 1;
author maintainer
#198 lore raw

PATCH 198/205 perf/core: Dont WARN() for impossible ring-buffer sizes

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Rutland <[email protected]>

commit 9dff0aa95a324e262ffb03f425d00e4751f3294e upstream.

The perf tool uses /proc/sys/kernel/perf_event_mlock_kb to determine how large its ringbuffer mmap should be. This can be configured to arbitrary values, which can be larger than the maximum possible allocation from kmalloc.

When this is configured to a suitably large value (e.g. thanks to the perf fuzzer), attempting to use perf record triggers a WARN_ON_ONCE() in __alloc_pages_nodemask():

   WARNING: CPU: 2 PID: 5666 at mm/page_alloc.c:4511 __alloc_pages_nodemask+0x3f8/0xbc8

Let's avoid this by checking that the requested allocation is possible before calling kzalloc.

---
 kernel/events/ring_buffer.c |    3 +++
 1 file changed, 3 insertions(+)
kernel/events/ring_buffer.c +3 -0
@@ -719,6 +719,9 @@ struct ring_buffer *rb_alloc(int nr_page
 	size = sizeof(struct ring_buffer);
 	size += nr_pages * sizeof(void *);
 
+	if (order_base_2(size) >= MAX_ORDER)
+		goto fail;
+
 	rb = kzalloc(size, GFP_KERNEL);
 	if (!rb)
 		goto fail;
#199 lore raw

Re: [PATCH 4.14 198/205] perf/core: Dont WARN() for impossible ring-buffer sizes

Quoted 21 lines · Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Rutland <[email protected]>

commit 9dff0aa95a324e262ffb03f425d00e4751f3294e upstream.

The perf tool uses /proc/sys/kernel/perf_event_mlock_kb to determine how
large its ringbuffer mmap should be. This can be configured to arbitrary
values, which can be larger than the maximum possible allocation from kmalloc.

When this is configured to a suitably large value (e.g. thanks to the perf fuzzer), attempting to use perf record triggers a WARN_ON_ONCE() in
__alloc_pages_nodemask():

Quoted 6 lines
   WARNING: CPU: 2 PID: 5666 at mm/page_alloc.c:4511
__alloc_pages_nodemask+0x3f8/0xbc8

Let's avoid this by checking that the requested allocation is possible
before calling kzalloc.

Hi,

Perf tool is broken for me in 4.14.99 (running in x86_64 VM), bisection points to this patch.

# perf top
Error:
Failed to mmap with 12 (Cannot allocate memory)
# perf trace
Cannot allocate memory
# strace -T -tt -f -y perf top
[...]
14:22:09.829544 openat(AT_FDCWD,
"/proc/sys/kernel/perf_event_mlock_kb", O_RDONLY) =
18</proc/sys/kernel/perf_event_mlock_kb> <0.000015>
14:22:09.829612 read(18</proc/sys/kernel/perf_event_mlock_kb>, "516\n",
64) = 4 <0.000011>
14:22:09.829655 close(18</proc/sys/kernel/perf_event_mlock_kb>) = 0
<0.000008>
14:22:09.829702 mmap(NULL, 528384, PROT_READ|PROT_WRITE, MAP_SHARED,
3<anon_inode:[perf_event]>, 0) = -1 ENOMEM (Cannot allocate memory)
<0.000015>
14:22:09.829763 write(2</dev/pts/0>, "Error:\n", 7) = 7 <0.000009>
14:22:09.829810 write(2</dev/pts/0>, "Failed to mmap with 12 (Cannot
a"..., 48) = 48 <0.000008>

Changing the patch like this fixes it...

- if (order_base_2(size) >= MAX_ORDER)
+ if (order_base_2(size) > MAX_ORDER)

-Tommi

Quoted 35 lines
Cc Alexander Shishkin <[email protected]>
Cc Arnaldo Carvalho de Melo <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Namhyung Kim <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Thomas Gleixner <[email protected]>

Link:
https://lkml.kernel.org/r/[email protected]

---
 kernel/events/ring_buffer.c |    3 +++
 1 file changed, 3 insertions(+)
kernel/events/ring_buffer.c +3 -0
@@ -719,6 +719,9 @@ struct ring_buffer *rb_alloc(int nr_page
 	size = sizeof(struct ring_buffer);
 	size += nr_pages * sizeof(void *);
 
+	if (order_base_2(size) >= MAX_ORDER)
+		goto fail;
+
 	rb = kzalloc(size, GFP_KERNEL);
 	if (!rb)
 		goto fail;
#200 lore raw

Re: Suspected SPAM - Re: [PATCH 4.14 198/205] perf/core: Dont WARN() for impossible ring-buffer sizes

Quoted 33 lines · Tommi T. (Nokia - FI/Espoo)
Quoted 25 lines · Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Rutland <[email protected]>

commit 9dff0aa95a324e262ffb03f425d00e4751f3294e upstream.

The perf tool uses /proc/sys/kernel/perf_event_mlock_kb to determine
how
large its ringbuffer mmap should be. This can be configured to arbitrary
values, which can be larger than the maximum possible allocation from
kmalloc.

When this is configured to a suitably large value (e.g. thanks to the
perf fuzzer), attempting to use perf record triggers a WARN_ON_ONCE()
in
__alloc_pages_nodemask():

Quoted 6 lines
   WARNING: CPU: 2 PID: 5666 at mm/page_alloc.c:4511
__alloc_pages_nodemask+0x3f8/0xbc8

Let's avoid this by checking that the requested allocation is possible
before calling kzalloc.

Quoted 5 lines

Hi,

Perf tool is broken for me in 4.14.99 (running in x86_64 VM), bisection
points to this patch.

... and I see there's a fix available:

https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?id=528871b456026e6127d95b1b2bd8e3a003dc1614

-Tommi

Quoted 37 lines
Quoted 37 lines
Cc Alexander Shishkin <[email protected]>
Cc Arnaldo Carvalho de Melo <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Namhyung Kim <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Thomas Gleixner <[email protected]>

Link:
https://lkml.kernel.org/r/[email protected]

---
 kernel/events/ring_buffer.c |    3 +++
 1 file changed, 3 insertions(+)
kernel/events/ring_buffer.c +3 -0
@@ -719,6 +719,9 @@ struct ring_buffer *rb_alloc(int nr_page
 	size = sizeof(struct ring_buffer);
 	size += nr_pages * sizeof(void *);
 
+	if (order_base_2(size) >= MAX_ORDER)
+		goto fail;
+
 	rb = kzalloc(size, GFP_KERNEL);
 	if (!rb)
 		goto fail;
#202 lore raw

Re: [PATCH 4.14 198/205] perf/core: Dont WARN() for impossible ring-buffer sizes

Quoted 4 lines · Peter Zijlstra

There is this queued:

http://lkml.kernel.org/r/[email protected]

I think it would likely have been better to just remove the test entirely, and instead just use __GFP_NOWARN.

This is an allocation essentially done for the user (as part of mmap(), and we handle the failure case fine. So it doesn't matter whether the warning is due to being out of memory or being a bad size, we shouldn't cause a kernel warning.

That said, I think we should *indepdendently* have some sane limit for "nr_pages", simply because we do that

        ...
        for (i = 0; i < nr_pages; i++) {
                rb->data_pages[i] = perf_mmap_alloc_page(cpu);

thing, which is really really expensive when "nr_pages" is in the thousands (or millions). The fact that we end up almost _accidentally_ limiting nr_pages simply because the kzmalloc() size is limited is I think a mistake.

So I'd rather see __GFP_NORWARN _and_ some kind of "don't allow crazy perf ring buffers" check that is separate and independent from any kmalloc issue..

For example, right now the maximum size you can allocate is (insanely) tied to the page size. With a bigger page size, not only do you get a bigger ring buffer anyway, but you also get more 'nr_pages', so it's basically quadratic behavior. Does that make sense? It doesn't sound that way to me.

Linus

author maintainer
#203 lore raw

PATCH 199/205 perf tests evsel-tp-sched: Fix bitwise operator

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Gustavo A. R. Silva <[email protected]>

commit 489338a717a0dfbbd5a3fabccf172b78f0ac9015 upstream.

Notice that the use of the bitwise OR operator '|' always leads to true in this particular case, which seems a bit suspicious due to the context in which this expression is being used.

Fix this by using bitwise AND operator '&' instead.

This bug was detected with the help of Coccinelle.

Cc Alexander Shishkin <[email protected]>
Cc Namhyung Kim <[email protected]>
Cc Peter Zijlstra <[email protected]>
Fixes 6a6cd11d4e57 ("perf test: Add test for the sched tracepoint format fields")
---
 tools/perf/tests/evsel-tp-sched.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/tests/evsel-tp-sched.c +1 -1
@@ -17,7 +17,7 @@ static int perf_evsel__test_field(struct
 		return -1;
 	}
 
-	is_signed = !!(field->flags | FIELD_IS_SIGNED);
+	is_signed = !!(field->flags & FIELD_IS_SIGNED);
 	if (should_be_signed && !is_signed) {
 		pr_debug("%s: \"%s\" signedness(%d) is wrong, should be %d\n",
 			 evsel->name, name, is_signed, should_be_signed);
author maintainer
#204 lore raw

PATCH 200/205 serial: fix race between flush_to_ldisc and tty_open

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

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

commit fedb5760648a291e949f2380d383b5b2d2749b5e upstream.

There still is a race window after the commit b027e2298bd588 ("tty: fix data race between tty_init_dev and flush of buf"), and we encountered this crash issue if receive_buf call comes before tty initialization completes in tty_open and tty->driver_data may be NULL.

CPU0 CPU1
---- ----
tty_open
tty_init_dev
tty_ldisc_unlock
schedule
flush_to_ldisc
receive_buf
tty_port_default_receive_buf
tty_ldisc_receive_buf

    n_tty_receive_buf_common
      __receive_buf
       uart_flush_chars
        uart_start
        /*tty->driver_data is NULL*/
                                   tty->ops->open
                                   /*init tty->driver_data*/

it can be fixed by extending ldisc semaphore lock in tty_init_dev to driver_data initialized completely after tty->ops->open(), but this will lead to get lock on one function and unlock in some other function, and hard to maintain, so fix this race only by checking tty->driver_data when receiving, and return if tty->driver_data is NULL, and n_tty_receive_buf_common maybe calls uart_unthrottle, so add the same check.

Because the tty layer knows nothing about the driver associated with the device, the tty layer can not do anything here, it is up to the tty driver itself to check for this type of race. Fix up the serial driver to correctly check to see if it is finished binding with the device when being called, and if not, abort the tty calls.

[Description and problem report and testing from Li RongQing, I rewrote the patch to be in the serial layer, not in the tty core - gregkh]

---
 drivers/tty/serial/serial_core.c |    6 ++++++
 1 file changed, 6 insertions(+)
drivers/tty/serial/serial_core.c +6 -0
@@ -143,6 +143,9 @@ static void uart_start(struct tty_struct
 	struct uart_port *port;
 	unsigned long flags;
 
+	if (!state)
+		return;
+
 	port = uart_port_lock(state, flags);
 	__uart_start(tty);
 	uart_port_unlock(port, flags);
@@ -2415,6 +2418,9 @@ static void uart_poll_put_char(struct tt
 	struct uart_state *state = drv->state + line;
 	struct uart_port *port;
 
+	if (!state)
+		return;
+
 	port = uart_port_ref(state);
 	if (!port)
 		return;
author maintainer
#205 lore raw

PATCH 201/205 serial: 8250_pci: Make PCI class test non fatal

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Shevchenko <[email protected]>

commit 824d17c57b0abbcb9128fb3f7327fae14761914b upstream.

As has been reported the National Instruments serial cards have broken PCI class.

The commit 7d8905d06405

("serial: 8250_pci: Enable device after we check black list")

made the PCI class check mandatory for the case when device is listed in a quirk list.

Make PCI class test non fatal to allow broken card be enumerated.

Fixes 7d8905d06405 ("serial: 8250_pci: Enable device after we check black list")
Cc stable <[email protected]>
---
 drivers/tty/serial/8250/8250_pci.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
drivers/tty/serial/8250/8250_pci.c +5 -4
@@ -3425,6 +3425,11 @@ static int
 serial_pci_guess_board(struct pci_dev *dev, struct pciserial_board *board)
 {
 	int num_iomem, num_port, first_port = -1, i;
+	int rc;
+
+	rc = serial_pci_is_class_communication(dev);
+	if (rc)
+		return rc;
 
 	/*
 	 * Should we try to make guesses for multiport serial devices later?
@@ -3652,10 +3657,6 @@ pciserial_init_one(struct pci_dev *dev,
 
 	board = &pci_boards[ent->driver_data];
 
-	rc = serial_pci_is_class_communication(dev);
-	if (rc)
-		return rc;
-
 	rc = serial_pci_is_blacklisted(dev);
 	if (rc)
 		return rc;
author maintainer
#206 lore raw

PATCH 202/205 nfsd4: fix cached replies to solo SEQUENCE compounds

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: J. Bruce Fields <[email protected]>

commit 085def3ade52f2ffe3e31f42e98c27dcc222dd37 upstream.

Currently our handling of 4.1+ requests without "cachethis" set is confusing and not quite correct.

Suppose a client sends a compound consisting of only a single SEQUENCE op, and it matches the seqid in a session slot (so it's a retry), but the previous request with that seqid did not have "cachethis" set.

The obvious thing to do might be to return NFS4ERR_RETRY_UNCACHED_REP, but the protocol only allows that to be returned on the op following the SEQUENCE, and there is no such op in this case.

The protocol permits us to cache replies even if the client didn't ask us to. And it's easy to do so in the case of solo SEQUENCE compounds.

So, when we get a solo SEQUENCE, we can either return the previously cached reply or NFSERR_SEQ_FALSE_RETRY if we notice it differs in some way from the original call.

Currently, we're returning a corrupt reply in the case a solo SEQUENCE matches a previous compound with more ops. This actually matters because the Linux client recently started doing this as a way to recover from lost replies to idempotent operations in the case the process doing the original reply was killed: in that case it's difficult to keep the original arguments around to do a real retry, and the client no longer cares what the result is anyway, but it would like to make sure that the slot's sequence id has been incremented, and the solo SEQUENCE assures
that: if the server never got the original reply, it will increment the sequence id. If it did get the original reply, it won't increment, and nothing else that about the reply really matters much. But we can at least attempt to return valid xdr!

---
 fs/nfsd/nfs4state.c |   20 +++++++++++++++-----
 fs/nfsd/state.h     |    1 +
 fs/nfsd/xdr4.h      |   13 +++++++++++--
 3 files changed, 27 insertions(+), 7 deletions(-)
fs/nfsd/nfs4state.c +15 -5
@@ -2331,14 +2331,16 @@ nfsd4_store_cache_entry(struct nfsd4_com
 
 	dprintk("--> %s slot %p\n", __func__, slot);
 
+	slot->sl_flags |= NFSD4_SLOT_INITIALIZED;
 	slot->sl_opcnt = resp->opcnt;
 	slot->sl_status = resp->cstate.status;
 
-	slot->sl_flags |= NFSD4_SLOT_INITIALIZED;
-	if (nfsd4_not_cached(resp)) {
-		slot->sl_datalen = 0;
+	if (!nfsd4_cache_this(resp)) {
+		slot->sl_flags &= ~NFSD4_SLOT_CACHED;
 		return;
 	}
+	slot->sl_flags |= NFSD4_SLOT_CACHED;
+
 	base = resp->cstate.data_offset;
 	slot->sl_datalen = buf->len - base;
 	if (read_bytes_from_xdr_buf(buf, base, slot->sl_data, slot->sl_datalen))
@@ -2365,8 +2367,16 @@ nfsd4_enc_sequence_replay(struct nfsd4_c
 	op = &args->ops[resp->opcnt - 1];
 	nfsd4_encode_operation(resp, op);
 
-	/* Return nfserr_retry_uncached_rep in next operation. */
-	if (args->opcnt > 1 && !(slot->sl_flags & NFSD4_SLOT_CACHETHIS)) {
+	if (slot->sl_flags & NFSD4_SLOT_CACHED)
+		return op->status;
+	if (args->opcnt == 1) {
+		/*
+		 * The original operation wasn't a solo sequence--we
+		 * always cache those--so this retry must not match the
+		 * original:
+		 */
+		op->status = nfserr_seq_false_retry;
+	} else {
 		op = &args->ops[resp->opcnt++];
 		op->status = nfserr_retry_uncached_rep;
 		nfsd4_encode_operation(resp, op);
fs/nfsd/state.h +1 -0
@@ -174,6 +174,7 @@ struct nfsd4_slot {
 #define NFSD4_SLOT_INUSE	(1 << 0)
 #define NFSD4_SLOT_CACHETHIS	(1 << 1)
 #define NFSD4_SLOT_INITIALIZED	(1 << 2)
+#define NFSD4_SLOT_CACHED	(1 << 3)
 	u8	sl_flags;
 	char	sl_data[];
 };
fs/nfsd/xdr4.h +11 -2
@@ -651,9 +651,18 @@ static inline bool nfsd4_is_solo_sequenc
 	return resp->opcnt == 1 && args->ops[0].opnum == OP_SEQUENCE;
 }
 
-static inline bool nfsd4_not_cached(struct nfsd4_compoundres *resp)
+/*
+ * The session reply cache only needs to cache replies that the client
+ * actually asked us to.  But it's almost free for us to cache compounds
+ * consisting of only a SEQUENCE op, so we may as well cache those too.
+ * Also, the protocol doesn't give us a convenient response in the case
+ * of a replay of a solo SEQUENCE op that wasn't cached
+ * (RETRY_UNCACHED_REP can only be returned in the second op of a
+ * compound).
+ */
+static inline bool nfsd4_cache_this(struct nfsd4_compoundres *resp)
 {
-	return !(resp->cstate.slot->sl_flags & NFSD4_SLOT_CACHETHIS)
+	return (resp->cstate.slot->sl_flags & NFSD4_SLOT_CACHETHIS)
 		|| nfsd4_is_solo_sequence(resp);
 }
 
author maintainer
#207 lore raw

PATCH 203/205 nfsd4: catch some false session retries

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: J. Bruce Fields <[email protected]>

commit 53da6a53e1d414e05759fa59b7032ee08f4e22d7 upstream.

The spec allows us to return NFS4ERR_SEQ_FALSE_RETRY if we notice that the client is making a call that matches a previous (slot, seqid) pair but that *isn't* actually a replay, because some detail of the call doesn't actually match the previous one.

Catching every such case is difficult, but we may as well catch a few easy ones. This also handles the case described in the previous patch, in a different way.

The spec does however require us to catch the case where the difference is in the rpc credentials. This prevents somebody from snooping another user's replies by fabricating retries.

(But the practical value of the attack is limited by the fact that the replies with the most sensitive data are READ replies, which are not normally cached.)

---
 fs/nfsd/nfs4state.c |   37 ++++++++++++++++++++++++++++++++++++-
 fs/nfsd/state.h     |    1 +
 2 files changed, 37 insertions(+), 1 deletion(-)
fs/nfsd/nfs4state.c +36 -1
@@ -1472,8 +1472,10 @@ free_session_slots(struct nfsd4_session
 {
 	int i;
 
-	for (i = 0; i < ses->se_fchannel.maxreqs; i++)
+	for (i = 0; i < ses->se_fchannel.maxreqs; i++) {
+		free_svc_cred(&ses->se_slots[i]->sl_cred);
 		kfree(ses->se_slots[i]);
+	}
 }
 
 /*
@@ -2334,6 +2336,8 @@ nfsd4_store_cache_entry(struct nfsd4_com
 	slot->sl_flags |= NFSD4_SLOT_INITIALIZED;
 	slot->sl_opcnt = resp->opcnt;
 	slot->sl_status = resp->cstate.status;
+	free_svc_cred(&slot->sl_cred);
+	copy_cred(&slot->sl_cred, &resp->rqstp->rq_cred);
 
 	if (!nfsd4_cache_this(resp)) {
 		slot->sl_flags &= ~NFSD4_SLOT_CACHED;
@@ -3040,6 +3044,34 @@ static bool nfsd4_request_too_big(struct
 	return xb->len > session->se_fchannel.maxreq_sz;
 }
 
+static bool replay_matches_cache(struct svc_rqst *rqstp,
+		 struct nfsd4_sequence *seq, struct nfsd4_slot *slot)
+{
+	struct nfsd4_compoundargs *argp = rqstp->rq_argp;
+
+	if ((bool)(slot->sl_flags & NFSD4_SLOT_CACHETHIS) !=
+	    (bool)seq->cachethis)
+		return false;
+	/*
+	 * If there's an error than the reply can have fewer ops than
+	 * the call.  But if we cached a reply with *more* ops than the
+	 * call you're sending us now, then this new call is clearly not
+	 * really a replay of the old one:
+	 */
+	if (slot->sl_opcnt < argp->opcnt)
+		return false;
+	/* This is the only check explicitly called by spec: */
+	if (!same_creds(&rqstp->rq_cred, &slot->sl_cred))
+		return false;
+	/*
+	 * There may be more comparisons we could actually do, but the
+	 * spec doesn't require us to catch every case where the calls
+	 * don't match (that would require caching the call as well as
+	 * the reply), so we don't bother.
+	 */
+	return true;
+}
+
 __be32
 nfsd4_sequence(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		union nfsd4_op_u *u)
@@ -3099,6 +3131,9 @@ nfsd4_sequence(struct svc_rqst *rqstp, s
 		status = nfserr_seq_misordered;
 		if (!(slot->sl_flags & NFSD4_SLOT_INITIALIZED))
 			goto out_put_session;
+		status = nfserr_seq_false_retry;
+		if (!replay_matches_cache(rqstp, seq, slot))
+			goto out_put_session;
 		cstate->slot = slot;
 		cstate->session = session;
 		cstate->clp = clp;
fs/nfsd/state.h +1 -0
@@ -169,6 +169,7 @@ static inline struct nfs4_delegation *de
 struct nfsd4_slot {
 	u32	sl_seqid;
 	__be32	sl_status;
+	struct svc_cred sl_cred;
 	u32	sl_datalen;
 	u16	sl_opcnt;
 #define NFSD4_SLOT_INUSE	(1 << 0)
author maintainer
#208 lore raw

PATCH 204/205 IB/hfi1: Add limit test for RC/UC send via loopback

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mike Marciniszyn <[email protected]>

commit 09ce351dff8e7636af0beb72cd4a86c3904a0500 upstream.

Fix potential memory corruption and panic in loopback for IB_WR_SEND variants.

The code blindly assumes the posted length will fit in the fetched rwqe, which is not a valid assumption.

Fix by adding a limit test, and triggering the appropriate send completion and putting the QP in an error state. This mimics the handling for non-loopback QPs.

Fixes 15703461533a ("IB/{hfi1, qib, rdmavt}: Move ruc_loopback to rdmavt")
Cc <[email protected]> #v4.20+
---
 drivers/infiniband/hw/hfi1/ruc.c    |    7 ++++++-
 drivers/infiniband/hw/qib/qib_ruc.c |    7 ++++++-
 2 files changed, 12 insertions(+), 2 deletions(-)
drivers/infiniband/hw/hfi1/ruc.c +6 -1
@@ -440,6 +440,8 @@ send:
 			goto op_err;
 		if (!ret)
 			goto rnr_nak;
+		if (wqe->length > qp->r_len)
+			goto inv_err;
 		break;
 
 	case IB_WR_RDMA_WRITE_WITH_IMM:
@@ -607,7 +609,10 @@ op_err:
 	goto err;
 
 inv_err:
-	send_status = IB_WC_REM_INV_REQ_ERR;
+	send_status =
+		sqp->ibqp.qp_type == IB_QPT_RC ?
+			IB_WC_REM_INV_REQ_ERR :
+			IB_WC_SUCCESS;
 	wc.status = IB_WC_LOC_QP_OP_ERR;
 	goto err;
 
drivers/infiniband/hw/qib/qib_ruc.c +6 -1
@@ -425,6 +425,8 @@ again:
 			goto op_err;
 		if (!ret)
 			goto rnr_nak;
+		if (wqe->length > qp->r_len)
+			goto inv_err;
 		break;
 
 	case IB_WR_RDMA_WRITE_WITH_IMM:
@@ -585,7 +587,10 @@ op_err:
 	goto err;
 
 inv_err:
-	send_status = IB_WC_REM_INV_REQ_ERR;
+	send_status =
+		sqp->ibqp.qp_type == IB_QPT_RC ?
+			IB_WC_REM_INV_REQ_ERR :
+			IB_WC_SUCCESS;
 	wc.status = IB_WC_LOC_QP_OP_ERR;
 	goto err;
 
author maintainer
#209 lore raw

PATCH 205/205 perf/x86/intel: Delay memory deallocation until x86_pmu_dead_cpu()

In reply to Greg Kroah-Hartman

4.14-stable review patch. If anyone has any objections, please let me know.

------------------

From: Peter Zijlstra <[email protected]>

commit 602cae04c4864bb3487dfe4c2126c8d9e7e1614a upstream.

intel_pmu_cpu_prepare() allocated memory for ->shared_regs among other members of struct cpu_hw_events. This memory is released in intel_pmu_cpu_dying() which is wrong. The counterpart of the intel_pmu_cpu_prepare() callback is x86_pmu_dead_cpu().

Otherwise if the CPU fails on the UP path between CPUHP_PERF_X86_PREPARE and CPUHP_AP_PERF_X86_STARTING then it won't release the memory but allocate new memory on the next attempt to online the CPU (leaking the old memory).
Also, if the CPU down path fails between CPUHP_AP_PERF_X86_STARTING and CPUHP_PERF_X86_PREPARE then the CPU will go back online but never allocate the memory that was released in x86_pmu_dying_cpu().

Make the memory allocation/free symmetrical in regard to the CPU hotplug notifier by moving the deallocation to intel_pmu_cpu_dead().

This started in commit:

a7e3ed1e47011 ("perf: Add support for supplementary event registers").

In principle the bug was introduced in v2.6.39 (!), but it will almost certainly not backport cleanly across the big CPU hotplug rewrite between v4.7-v4.15...

[ bigeasy: Added patch description. ]
[ mingo: Added backporting guidance. ]

[ He Zhe: Fixes conflict caused by missing disable_counter_freeze which is
introduced since v4.20 af3bdb991a5cb. ]

---
 arch/x86/events/intel/core.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
arch/x86/events/intel/core.c +8 -2
@@ -3420,6 +3420,11 @@ static void free_excl_cntrs(int cpu)
 
 static void intel_pmu_cpu_dying(int cpu)
 {
+	fini_debug_store_on_cpu(cpu);
+}
+
+static void intel_pmu_cpu_dead(int cpu)
+{
 	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
 	struct intel_shared_regs *pc;
 
@@ -3431,8 +3436,6 @@ static void intel_pmu_cpu_dying(int cpu)
 	}
 
 	free_excl_cntrs(cpu);
-
-	fini_debug_store_on_cpu(cpu);
 }
 
 static void intel_pmu_sched_task(struct perf_event_context *ctx,
@@ -3521,6 +3524,7 @@ static __initconst const struct x86_pmu
 	.cpu_prepare		= intel_pmu_cpu_prepare,
 	.cpu_starting		= intel_pmu_cpu_starting,
 	.cpu_dying		= intel_pmu_cpu_dying,
+	.cpu_dead		= intel_pmu_cpu_dead,
 };
 
 static struct attribute *intel_pmu_attrs[];
@@ -3560,6 +3564,8 @@ static __initconst const struct x86_pmu
 	.cpu_prepare		= intel_pmu_cpu_prepare,
 	.cpu_starting		= intel_pmu_cpu_starting,
 	.cpu_dying		= intel_pmu_cpu_dying,
+	.cpu_dead		= intel_pmu_cpu_dead,
+
 	.guest_get_msrs		= intel_guest_get_msrs,
 	.sched_task		= intel_pmu_sched_task,
 };
#210 lore raw
In reply to Greg Kroah-Hartman
Quoted 10 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 4.14.99 release. There are 205 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed Feb 13 14:17:19 UTC 2019. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 9 lines
        https://www.kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.14.99-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.14.y
and the diffstat can be found below.

thanks,

greg k-h

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

Summary
------------------------------------------------------------------------

kernel: 4.14.99-rc1
git repo: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
git branch: linux-4.14.y
git commit: 151d6387b56840f9daaf468ec2212d2ac9ccc047
git describe: v4.14.98-206-g151d6387b568
Test details: https://qa-reports.linaro.org/lkft/linux-stable-rc-4.14-oe/build/v4.14.98-206-g151d6387b568

No regressions (compared to build v4.14.98)

No fixes (compared to build v4.14.98)

Ran 22227 total tests in the following environments and test suites.

Environments
--------------
- dragonboard-410c - arm64
- hi6220-hikey - arm64
- i386
- juno-r2 - arm64
- qemu_arm
- qemu_arm64
- qemu_i386
- qemu_x86_64
- x15 - arm
- x86_64

Test Suites
-----------
* boot
* install-android-platform-tools-r2600
* kselftest
* libhugetlbfs
* ltp-cap_bounds-tests
* ltp-containers-tests
* ltp-cpuhotplug-tests
* ltp-cve-tests
* ltp-fcntl-locktests-tests
* ltp-filecaps-tests
* ltp-fs-tests
* ltp-fs_bind-tests
* ltp-fs_perms_simple-tests
* ltp-fsx-tests
* ltp-hugetlb-tests
* ltp-io-tests
* ltp-ipc-tests
* ltp-math-tests
* ltp-mm-tests
* ltp-nptl-tests
* ltp-pty-tests
* ltp-sched-tests
* ltp-securebits-tests
* ltp-syscalls-tests
* spectre-meltdown-checker-test
* ltp-timers-tests
* ltp-open-posix-tests
* kselftest-vsyscall-mode-native
* kselftest-vsyscall-mode-none

Signature · 2 lines
Linaro LKFT
https://lkft.linaro.org
maintainer
#211 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 4.14.99 release. There are 205 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed Feb 13 14:17:19 UTC 2019. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.14.99-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.14.y
and the diffstat can be found below.

thanks,

greg k-h

All tests are passing for Tegra ...

Test results for stable-v4.14:

    8 builds:	8 pass, 0 fail
    16 boots:	16 pass, 0 fail
    14 tests:	14 pass, 0 fail

Linux version: 4.14.99-rc1-g151d638
Boards tested: tegra124-jetson-tk1, tegra20-ventana,

                tegra210-p2371-2180, tegra30-cardhu-a04

BTW, I did not seem to receive a 'stable review' for v4.4.175-rc1. I am not sure if it is just me, but that last one I see in the archive for v4.4 [0] is v4.4-174-rc1.

Cheers
Jon

[0] https://www.spinics.net/lists/stable/

Signature · 1 line
nvpublic
author maintainer
#212 lore raw
Quoted 21 lines · Jon Hunter
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 4.14.99 release. There are 205 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed Feb 13 14:17:19 UTC 2019. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.14.99-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.14.y
and the diffstat can be found below.

thanks,

greg k-h

All tests are passing for Tegra ...

Quoted 8 lines

Test results for stable-v4.14:

    8 builds:	8 pass, 0 fail
    16 boots:	16 pass, 0 fail
    14 tests:	14 pass, 0 fail

Linux version: 4.14.99-rc1-g151d638
Boards tested: tegra124-jetson-tk1, tegra20-ventana,

                tegra210-p2371-2180, tegra30-cardhu-a04

Thanks for testing 2 of these.

Quoted 3 lines

BTW, I did not seem to receive a 'stable review' for v4.4.175-rc1. I am not sure if it is just me, but that last one I see in the archive for v4.4 [0] is v4.4-174-rc1.

No, I didn't do a "real" 4.4-rc just yet, I wanted to get these 4 kernels out first.

thanks,

greg k-h

#213 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 4.14.99 release. There are 205 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed Feb 13 14:17:19 UTC 2019. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 9 lines
	https://www.kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.14.99-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.14.y
and the diffstat can be found below.

thanks,

greg k-h

Compiled and booted on my test system. No dmesg regressions.

thanks,
-- Shuah

#214 lore raw
In reply to Greg Kroah-Hartman
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 4.14.99 release. There are 205 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed Feb 13 14:17:19 UTC 2019. Anything received after that time might be too late.

Build results:

	total: 172 pass: 172 fail: 0
Qemu test results:
	total: 328 pass: 328 fail: 0

Guenter