Skip to contentSkip to conversation

ACPI: NUMA: Use correct type for printing addresses on i386-PAE

PATCH New Sasha Levin 1 participant 16 messages Feb 9, 2019

Subsystems (5)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 1/16 ACPI: NUMA: Use correct type for printing addresses on i386-PAE

From: Chao Fan <[email protected]>

[ Upstream commit b9ced18acf68dffebe6888c7ec765a2b1db7a039 ]

The addresses of NUMA nodes are not printed correctly on i386-PAE which is misleading.

Here is a debian9-32bit with PAE in a QEMU guest having more than 4G of memory:

qemu-system-i386 \
-hda /var/lib/libvirt/images/debian32.qcow2 \
-m 5G \
-enable-kvm \
-smp 10 \
-numa node,mem=512M,nodeid=0,cpus=0 \
-numa node,mem=512M,nodeid=1,cpus=1 \
-numa node,mem=512M,nodeid=2,cpus=2 \
-numa node,mem=512M,nodeid=3,cpus=3 \
-numa node,mem=512M,nodeid=4,cpus=4 \
-numa node,mem=512M,nodeid=5,cpus=5 \
-numa node,mem=512M,nodeid=6,cpus=6 \
-numa node,mem=512M,nodeid=7,cpus=7 \
-numa node,mem=512M,nodeid=8,cpus=8 \
-numa node,mem=512M,nodeid=9,cpus=9 \
-serial stdio

Because of the wrong value type, it prints as below:

[    0.021049] ACPI: SRAT Memory (0x0 length 0xa0000) in proximity domain 0 enabled
[    0.021740] ACPI: SRAT Memory (0x100000 length 0x1ff00000) in proximity domain 0 enabled
[    0.022425] ACPI: SRAT Memory (0x20000000 length 0x20000000) in proximity domain 1 enabled
[    0.023092] ACPI: SRAT Memory (0x40000000 length 0x20000000) in proximity domain 2 enabled
[    0.023764] ACPI: SRAT Memory (0x60000000 length 0x20000000) in proximity domain 3 enabled
[    0.024431] ACPI: SRAT Memory (0x80000000 length 0x20000000) in proximity domain 4 enabled
[    0.025104] ACPI: SRAT Memory (0xa0000000 length 0x20000000) in proximity domain 5 enabled
[    0.025791] ACPI: SRAT Memory (0x0 length 0x20000000) in proximity domain 6 enabled
[    0.026412] ACPI: SRAT Memory (0x20000000 length 0x20000000) in proximity domain 7 enabled
[    0.027118] ACPI: SRAT Memory (0x40000000 length 0x20000000) in proximity domain 8 enabled
[    0.027802] ACPI: SRAT Memory (0x60000000 length 0x20000000) in proximity domain 9 enabled

The upper half of the start address of the NUMA domains between 6 and 9 inclusive was cut, so the printed values are incorrect.

Fix the value type, to get the correct values in the log as follows:

[    0.023698] ACPI: SRAT Memory (0x0 length 0xa0000) in proximity domain 0 enabled
[    0.024325] ACPI: SRAT Memory (0x100000 length 0x1ff00000) in proximity domain 0 enabled
[    0.024981] ACPI: SRAT Memory (0x20000000 length 0x20000000) in proximity domain 1 enabled
[    0.025659] ACPI: SRAT Memory (0x40000000 length 0x20000000) in proximity domain 2 enabled
[    0.026317] ACPI: SRAT Memory (0x60000000 length 0x20000000) in proximity domain 3 enabled
[    0.026980] ACPI: SRAT Memory (0x80000000 length 0x20000000) in proximity domain 4 enabled
[    0.027635] ACPI: SRAT Memory (0xa0000000 length 0x20000000) in proximity domain 5 enabled
[    0.028311] ACPI: SRAT Memory (0x100000000 length 0x20000000) in proximity domain 6 enabled
[    0.028985] ACPI: SRAT Memory (0x120000000 length 0x20000000) in proximity domain 7 enabled
[    0.029667] ACPI: SRAT Memory (0x140000000 length 0x20000000) in proximity domain 8 enabled
[    0.030334] ACPI: SRAT Memory (0x160000000 length 0x20000000) in proximity domain 9 enabled
[ rjw: Subject & changelog ]
---
 drivers/acpi/numa.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/acpi/numa.c +3 -3
index 8fb74d9011da..a7907b58562a 100644
@@ -147,9 +147,9 @@ acpi_table_print_srat_entry(struct acpi_subtable_header *header)
 		{
 			struct acpi_srat_mem_affinity *p =
 			    (struct acpi_srat_mem_affinity *)header;
-			pr_debug("SRAT Memory (0x%lx length 0x%lx) in proximity domain %d %s%s%s\n",
-				 (unsigned long)p->base_address,
-				 (unsigned long)p->length,
+			pr_debug("SRAT Memory (0x%llx length 0x%llx) in proximity domain %d %s%s%s\n",
+				 (unsigned long long)p->base_address,
+				 (unsigned long long)p->length,
 				 p->proximity_domain,
 				 (p->flags & ACPI_SRAT_MEM_ENABLED) ?
 				 "enabled" : "disabled",
Signature · 1 line
2.19.1
author maintainer
#2 lore raw

PATCH 3/16 cpufreq: check if policy is inactive early in __cpufreq_get()

From: Sudeep Holla <[email protected]>

[ Upstream commit 2f66196208c98b3d1b4294edffb2c5a8197be899 ]

cpuinfo_cur_freq gets current CPU frequency as detected by hardware while scaling_cur_freq last known CPU frequency. Some platforms may not allow checking the CPU frequency of an offline CPU or the associated resources may have been released via cpufreq_exit when the CPU gets offlined, in which case the policy would have been invalidated already. If we attempt to get current frequency from the hardware, it may result in hang or crash.

For example on Juno, I see:

Unable to handle kernel NULL pointer dereference at virtual address 0000000000000188
[0000000000000188] pgd=0000000000000000
Internal error: Oops: 96000004 [#1] PREEMPT SMP
Modules linked in:
CPU: 5 PID: 4202 Comm: cat Not tainted 4.20.0-08251-ga0f2c0318a15-dirty #87
Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform
pstate: 40000005 (nZcv daif -PAN -UAO)
pc : scmi_cpufreq_get_rate+0x34/0xb0
lr : scmi_cpufreq_get_rate+0x34/0xb0
Call trace:
 scmi_cpufreq_get_rate+0x34/0xb0
 __cpufreq_get+0x34/0xc0
 show_cpuinfo_cur_freq+0x24/0x78
 show+0x40/0x60
 sysfs_kf_seq_show+0xc0/0x148
 kernfs_seq_show+0x44/0x50
 seq_read+0xd4/0x480
 kernfs_fop_read+0x15c/0x208
 __vfs_read+0x60/0x188
 vfs_read+0x94/0x150
 ksys_read+0x6c/0xd8
 __arm64_sys_read+0x24/0x30
 el0_svc_common+0x78/0x100
 el0_svc_handler+0x38/0x78
 el0_svc+0x8/0xc
---[ end trace 3d1024e58f77f6b2 ]---

So fix the issue by checking if the policy is invalid early in __cpufreq_get before attempting to get the current frequency.

---
 drivers/cpufreq/cpufreq.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)
drivers/cpufreq/cpufreq.c +4 -8
index 93754300cb57..66c2790dcc5f 100644
@@ -1523,17 +1523,16 @@ static unsigned int __cpufreq_get(struct cpufreq_policy *policy)
 {
 	unsigned int ret_freq = 0;
 
-	if (!cpufreq_driver->get)
+	if (unlikely(policy_is_inactive(policy)) || !cpufreq_driver->get)
 		return ret_freq;
 
 	ret_freq = cpufreq_driver->get(policy->cpu);
 
 	/*
-	 * Updating inactive policies is invalid, so avoid doing that.  Also
-	 * if fast frequency switching is used with the given policy, the check
+	 * If fast frequency switching is used with the given policy, the check
 	 * against policy->cur is pointless, so skip it in that case too.
 	 */
-	if (unlikely(policy_is_inactive(policy)) || policy->fast_switch_enabled)
+	if (policy->fast_switch_enabled)
 		return ret_freq;
 
 	if (ret_freq && policy->cur &&
@@ -1562,10 +1561,7 @@ unsigned int cpufreq_get(unsigned int cpu)
 
 	if (policy) {
 		down_read(&policy->rwsem);
-
-		if (!policy_is_inactive(policy))
-			ret_freq = __cpufreq_get(policy);
-
+		ret_freq = __cpufreq_get(policy);
 		up_read(&policy->rwsem);
 
 		cpufreq_cpu_put(policy);
Signature · 1 line
2.19.1
author maintainer
#5 lore raw

PATCH 6/16 drm/bridge: tc358767: fix initial DP0/1_SRCCTRL value

In reply to Sasha Levin

From: Tomi Valkeinen <[email protected]>

[ Upstream commit 9a63bd6fe1b5590ffa42ae2ed22ee21363293e31 ]

Initially DP0_SRCCTRL is set to a static value which includes DP0_SRCCTRL_LANES_2 and DP0_SRCCTRL_BW27, even when only 1 lane of 1.62Gbps speed is used. DP1_SRCCTRL is configured to a magic number.

This patch changes the configuration as follows:

Configure DP0_SRCCTRL by using tc_srcctrl() which provides the correct value.

DP1_SRCCTRL needs two bits to be set to the same value as DP0_SRCCTRL:
SSCG and BW27. All other bits can be zero.

---
 drivers/gpu/drm/bridge/tc358767.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
drivers/gpu/drm/bridge/tc358767.c +5 -6
index acac2c1769ad..24bb6bfa36f3 100644
@@ -834,12 +834,11 @@ static int tc_main_link_setup(struct tc_data *tc)
 	if (!tc->mode)
 		return -EINVAL;
 
-	/* from excel file - DP0_SrcCtrl */
-	tc_write(DP0_SRCCTRL, DP0_SRCCTRL_SCRMBLDIS | DP0_SRCCTRL_EN810B |
-		 DP0_SRCCTRL_LANESKEW | DP0_SRCCTRL_LANES_2 |
-		 DP0_SRCCTRL_BW27 | DP0_SRCCTRL_AUTOCORRECT);
-	/* from excel file - DP1_SrcCtrl */
-	tc_write(DP1_SRCCTRL, 0x00003083);
+	tc_write(DP0_SRCCTRL, tc_srcctrl(tc));
+	/* SSCG and BW27 on DP1 must be set to the same as on DP0 */
+	tc_write(DP1_SRCCTRL,
+		 (tc->link.spread ? DP0_SRCCTRL_SSCG : 0) |
+		 ((tc->link.base.rate != 162000) ? DP0_SRCCTRL_BW27 : 0));
 
 	rate = clk_get_rate(tc->refclk);
 	switch (rate) {
Signature · 1 line
2.19.1
author maintainer
#8 lore raw

PATCH 9/16 nvme-pci: use the same attributes when freeing host_mem_desc_bufs.

In reply to Sasha Levin

From: Liviu Dudau <[email protected]>

[ Upstream commit cc667f6d5de023ee131e96bb88e5cddca23272bd ]

When using HMB the PCIe host driver allocates host_mem_desc_bufs using dma_alloc_attrs() but frees them using dma_free_coherent(). Use the correct dma_free_attrs() function to free the buffers.

Signed-off-by Liviu Dudau <liviu at dudau.co.uk>
Signed-off-by Christoph Hellwig <hch at lst.de>
Signed-off-by Sasha Levin <sashal at kernel.org>
---
 drivers/nvme/host/pci.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/nvme/host/pci.c +6 -4
index afb99876fa9e..06355ca832db 100644
@@ -1624,8 +1624,9 @@ static void nvme_free_host_mem(struct nvme_dev *dev)
 		struct nvme_host_mem_buf_desc *desc = &dev->host_mem_descs[i];
 		size_t size = le32_to_cpu(desc->size) * dev->ctrl.page_size;
 
-		dma_free_coherent(dev->dev, size, dev->host_mem_desc_bufs[i],
-				le64_to_cpu(desc->addr));
+		dma_free_attrs(dev->dev, size, dev->host_mem_desc_bufs[i],
+			       le64_to_cpu(desc->addr),
+			       DMA_ATTR_NO_KERNEL_MAPPING | DMA_ATTR_NO_WARN);
 	}
 
 	kfree(dev->host_mem_desc_bufs);
@@ -1691,8 +1692,9 @@ static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred,
 	while (--i >= 0) {
 		size_t size = le32_to_cpu(descs[i].size) * dev->ctrl.page_size;
 
-		dma_free_coherent(dev->dev, size, bufs[i],
-				le64_to_cpu(descs[i].addr));
+		dma_free_attrs(dev->dev, size, bufs[i],
+			       le64_to_cpu(descs[i].addr),
+			       DMA_ATTR_NO_KERNEL_MAPPING | DMA_ATTR_NO_WARN);
 	}
 
 	kfree(bufs);
Signature · 1 line
2.19.1
author maintainer
#9 lore raw

PATCH 10/16 ARM: dts: da850-evm: Correct the sound card name

In reply to Sasha Levin

From: Peter Ujfalusi <[email protected]>

[ Upstream commit 7fca69d4e43fa1ae9cb4f652772c132dc5a659c6 ]

To avoid the following error:
asoc-simple-card sound: ASoC: Failed to create card debugfs directory

Which is because the card name contains '/' character, which can not be used in file or directory names.

---
 arch/arm/boot/dts/da850-evm.dts | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/da850-evm.dts +1 -1
index c75507922f7d..f5902bd1a972 100644
@@ -169,7 +169,7 @@
 
 	sound {
 		compatible = "simple-audio-card";
-		simple-audio-card,name = "DA850/OMAP-L138 EVM";
+		simple-audio-card,name = "DA850-OMAPL138 EVM";
 		simple-audio-card,widgets =
 			"Line", "Line In",
 			"Line", "Line Out";
Signature · 1 line
2.19.1
author maintainer
#10 lore raw

PATCH 11/16 ARM: dts: da850-lcdk: Correct the sound card name

In reply to Sasha Levin

From: Peter Ujfalusi <[email protected]>

[ Upstream commit c25748acc5c20786ecb7518bfeae8fcef93472d6 ]

To avoid the following error:
asoc-simple-card sound: ASoC: Failed to create card debugfs directory

Which is because the card name contains '/' character, which can not be used in file or directory names.

---
 arch/arm/boot/dts/da850-lcdk.dts | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/da850-lcdk.dts +1 -1
index a0f0916156e6..c9d4cb212b72 100644
@@ -28,7 +28,7 @@
 
 	sound {
 		compatible = "simple-audio-card";
-		simple-audio-card,name = "DA850/OMAP-L138 LCDK";
+		simple-audio-card,name = "DA850-OMAPL138 LCDK";
 		simple-audio-card,widgets =
 			"Line", "Line In",
 			"Line", "Line Out";
Signature · 1 line
2.19.1
author maintainer
#11 lore raw

PATCH 12/16 ARM: dts: kirkwood: Fix polarity of GPIO fan lines

In reply to Sasha Levin

From: Linus Walleij <[email protected]>

[ Upstream commit b5f034845e70916fd33e172fad5ad530a29c10ab ]

These two lines are active high, not active low. The bug was found when we changed the kernel to respect the polarity defined in the device tree.

Fixes 1b90e06b1429 ("ARM: kirkwood: Use devicetree to define DNS-32[05] fan")
Cc Jamie Lentin <[email protected]>
Cc Guenter Roeck <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Andrew Lunn <[email protected]>
Cc Gregory Clement <[email protected]>
Cc Sebastian Hesselbarth <[email protected]>
Cc Julien D'Ascenzio <[email protected]>
---
 arch/arm/boot/dts/kirkwood-dnskw.dtsi | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm/boot/dts/kirkwood-dnskw.dtsi +2 -2
index cbaf06f2f78e..eb917462b219 100644
@@ -36,8 +36,8 @@
 		compatible = "gpio-fan";
 		pinctrl-0 = <&pmx_fan_high_speed &pmx_fan_low_speed>;
 		pinctrl-names = "default";
-		gpios = <&gpio1 14 GPIO_ACTIVE_LOW
-			 &gpio1 13 GPIO_ACTIVE_LOW>;
+		gpios = <&gpio1 14 GPIO_ACTIVE_HIGH
+			 &gpio1 13 GPIO_ACTIVE_HIGH>;
 		gpio-fan,speed-map = <0    0
 				      3000 1
 				      6000 2>;
Signature · 1 line
2.19.1
author maintainer
#13 lore raw

PATCH 14/16 drm/nouveau: Don't disable polling in fallback mode

In reply to Sasha Levin

From: Takashi Iwai <[email protected]>

[ Upstream commit 118780066e30c34de3d9349710b51780bfa0ba83 ]

When a fan is controlled via linear fallback without cstate, we shouldn't stop polling. Otherwise it won't be adjusted again and keeps running at an initial crazy pace.

Fixes 800efb4c2857 ("drm/nouveau/drm/therm/fan: add a fallback if no fan control is specified in the vbios")

Bugzilla: https://bugzilla.suse.com/show_bug.cgi?id=1103356
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107447

---
 drivers/gpu/drm/nouveau/nvkm/subdev/therm/base.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/gpu/drm/nouveau/nvkm/subdev/therm/base.c +4 -3
index 952a7cb0a59a..692d4d96766a 100644
@@ -131,11 +131,12 @@ nvkm_therm_update(struct nvkm_therm *therm, int mode)
 			duty = nvkm_therm_update_linear(therm);
 			break;
 		case NVBIOS_THERM_FAN_OTHER:
-			if (therm->cstate)
+			if (therm->cstate) {
 				duty = therm->cstate;
-			else
+				poll = false;
+			} else {
 				duty = nvkm_therm_update_linear_fallback(therm);
-			poll = false;
+			}
 			break;
 		}
 		immd = false;
Signature · 6 lines
2.19.1

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
author maintainer
#14 lore raw

PATCH 15/16 drm/nouveau/falcon: avoid touching registers if engine is off

In reply to Sasha Levin

From: Ilia Mirkin <[email protected]>

[ Upstream commit a5176a4cb85bb6213daadf691097cf411da35df2 ]

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108980

---
 drivers/gpu/drm/nouveau/nvkm/engine/falcon.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/gpu/drm/nouveau/nvkm/engine/falcon.c +5 -2
index 2e7b4e2105ef..62cb376e2c01 100644
@@ -22,6 +22,7 @@
 #include <engine/falcon.h>
 
 #include <core/gpuobj.h>
+#include <subdev/mc.h>
 #include <subdev/timer.h>
 #include <engine/fifo.h>
 
@@ -107,8 +108,10 @@ nvkm_falcon_fini(struct nvkm_engine *engine, bool suspend)
 		}
 	}
 
-	nvkm_mask(device, base + 0x048, 0x00000003, 0x00000000);
-	nvkm_wr32(device, base + 0x014, 0xffffffff);
+	if (nvkm_mc_enabled(device, engine->subdev.index)) {
+		nvkm_mask(device, base + 0x048, 0x00000003, 0x00000000);
+		nvkm_wr32(device, base + 0x014, 0xffffffff);
+	}
 	return 0;
 }
 
Signature · 6 lines
2.19.1

_______________________________________________
Nouveau mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/nouveau
author maintainer
#15 lore raw

PATCH 16/16 cifs: Limit memory used by lock request calls to a page

In reply to Sasha Levin

From: Ross Lagerwall <[email protected]>

[ Upstream commit 92a8109e4d3a34fb6b115c9098b51767dc933444 ]

The code tries to allocate a contiguous buffer with a size supplied by the server (maxBuf). This could fail if memory is fragmented since it results in high order allocations for commonly used server implementations. It is also wasteful since there are probably few locks in the usual case. Limit the buffer to be no larger than a page to avoid memory allocation failures due to fragmentation.

---
 fs/cifs/file.c     | 8 ++++++++
 fs/cifs/smb2file.c | 4 ++++
 2 files changed, 12 insertions(+)
fs/cifs/file.c +8 -0
index 1e176e11dbfa..852d7d1dcbbd 100644
@@ -1128,6 +1128,10 @@ cifs_push_mandatory_locks(struct cifsFileInfo *cfile)
 		return -EINVAL;
 	}
 
+	BUILD_BUG_ON(sizeof(struct smb_hdr) + sizeof(LOCKING_ANDX_RANGE) >
+		     PAGE_SIZE);
+	max_buf = min_t(unsigned int, max_buf - sizeof(struct smb_hdr),
+			PAGE_SIZE);
 	max_num = (max_buf - sizeof(struct smb_hdr)) /
 						sizeof(LOCKING_ANDX_RANGE);
 	buf = kcalloc(max_num, sizeof(LOCKING_ANDX_RANGE), GFP_KERNEL);
@@ -1466,6 +1470,10 @@ cifs_unlock_range(struct cifsFileInfo *cfile, struct file_lock *flock,
 	if (max_buf < (sizeof(struct smb_hdr) + sizeof(LOCKING_ANDX_RANGE)))
 		return -EINVAL;
 
+	BUILD_BUG_ON(sizeof(struct smb_hdr) + sizeof(LOCKING_ANDX_RANGE) >
+		     PAGE_SIZE);
+	max_buf = min_t(unsigned int, max_buf - sizeof(struct smb_hdr),
+			PAGE_SIZE);
 	max_num = (max_buf - sizeof(struct smb_hdr)) /
 						sizeof(LOCKING_ANDX_RANGE);
 	buf = kcalloc(max_num, sizeof(LOCKING_ANDX_RANGE), GFP_KERNEL);
fs/cifs/smb2file.c +4 -0
index 79078533f807..1add404618f0 100644
@@ -130,6 +130,8 @@ smb2_unlock_range(struct cifsFileInfo *cfile, struct file_lock *flock,
 	if (max_buf < sizeof(struct smb2_lock_element))
 		return -EINVAL;
 
+	BUILD_BUG_ON(sizeof(struct smb2_lock_element) > PAGE_SIZE);
+	max_buf = min_t(unsigned int, max_buf, PAGE_SIZE);
 	max_num = max_buf / sizeof(struct smb2_lock_element);
 	buf = kcalloc(max_num, sizeof(struct smb2_lock_element), GFP_KERNEL);
 	if (!buf)
@@ -266,6 +268,8 @@ smb2_push_mandatory_locks(struct cifsFileInfo *cfile)
 		return -EINVAL;
 	}
 
+	BUILD_BUG_ON(sizeof(struct smb2_lock_element) > PAGE_SIZE);
+	max_buf = min_t(unsigned int, max_buf, PAGE_SIZE);
 	max_num = max_buf / sizeof(struct smb2_lock_element);
 	buf = kcalloc(max_num, sizeof(struct smb2_lock_element), GFP_KERNEL);
 	if (!buf) {
Signature · 1 line
2.19.1
author maintainer
#3 lore raw

PATCH 4/16 drm/bridge: tc358767: add defines for DP1_SRCCTRL & PHY_2LANE

In reply to Sasha Levin

From: Tomi Valkeinen <[email protected]>

[ Upstream commit adf4109896bbee27fd2ac3b48d22d6a0062fe517 ]

DP1_SRCCTRL register and PHY_2LANE field did not have matching defines. Add these.

---
 drivers/gpu/drm/bridge/tc358767.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/gpu/drm/bridge/tc358767.c +7 -3
index 8636e7eeb731..e697c7c6ca52 100644
@@ -140,6 +140,8 @@
 #define DP0_LTLOOPCTRL		0x06d8
 #define DP0_SNKLTCTRL		0x06e4
 
+#define DP1_SRCCTRL		0x07a0
+
 /* PHY */
 #define DP_PHY_CTRL		0x0800
 #define DP_PHY_RST			BIT(28)  /* DP PHY Global Soft Reset */
@@ -148,6 +150,7 @@
 #define PHY_M1_RST			BIT(12)  /* Reset PHY1 Main Channel */
 #define PHY_RDY				BIT(16)  /* PHY Main Channels Ready */
 #define PHY_M0_RST			BIT(8)   /* Reset PHY0 Main Channel */
+#define PHY_2LANE			BIT(2)   /* PHY Enable 2 lanes */
 #define PHY_A0_EN			BIT(1)   /* PHY Aux Channel0 Enable */
 #define PHY_M0_EN			BIT(0)   /* PHY Main Channel0 Enable */
 
@@ -562,7 +565,7 @@ static int tc_aux_link_setup(struct tc_data *tc)
 	value |= SYSCLK_SEL_LSCLK | LSCLK_DIV_2;
 	tc_write(SYS_PLLPARAM, value);
 
-	tc_write(DP_PHY_CTRL, BGREN | PWR_SW_EN | BIT(2) | PHY_A0_EN);
+	tc_write(DP_PHY_CTRL, BGREN | PWR_SW_EN | PHY_2LANE | PHY_A0_EN);
 
 	/*
 	 * Initially PLLs are in bypass. Force PLL parameter update,
@@ -832,7 +835,7 @@ static int tc_main_link_setup(struct tc_data *tc)
 		 DP0_SRCCTRL_LANESKEW | DP0_SRCCTRL_LANES_2 |
 		 DP0_SRCCTRL_BW27 | DP0_SRCCTRL_AUTOCORRECT);
 	/* from excel file - DP1_SrcCtrl */
-	tc_write(0x07a0, 0x00003083);
+	tc_write(DP1_SRCCTRL, 0x00003083);
 
 	rate = clk_get_rate(tc->refclk);
 	switch (rate) {
@@ -853,8 +856,9 @@ static int tc_main_link_setup(struct tc_data *tc)
 	}
 	value |= SYSCLK_SEL_LSCLK | LSCLK_DIV_2;
 	tc_write(SYS_PLLPARAM, value);
+
 	/* Setup Main Link */
-	dp_phy_ctrl = BGREN | PWR_SW_EN | BIT(2) | PHY_A0_EN |  PHY_M0_EN;
+	dp_phy_ctrl = BGREN | PWR_SW_EN | PHY_2LANE | PHY_A0_EN |  PHY_M0_EN;
 	tc_write(DP_PHY_CTRL, dp_phy_ctrl);
 	msleep(100);
 
Signature · 6 lines
2.19.1

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
author maintainer
#4 lore raw

PATCH 5/16 drm/bridge: tc358767: fix single lane configuration

In reply to Sasha Levin

From: Tomi Valkeinen <[email protected]>

[ Upstream commit 4d9d54a730434cc068dd3515ba6116697196f77b ]

PHY_2LANE bit is always set in DP_PHY_CTRL, breaking 1 lane use.

Set PHY_2LANE only when 2 lanes are used.

---
 drivers/gpu/drm/bridge/tc358767.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
drivers/gpu/drm/bridge/tc358767.c +8 -2
index e697c7c6ca52..acac2c1769ad 100644
@@ -541,6 +541,7 @@ static int tc_aux_link_setup(struct tc_data *tc)
 	unsigned long rate;
 	u32 value;
 	int ret;
+	u32 dp_phy_ctrl;
 
 	rate = clk_get_rate(tc->refclk);
 	switch (rate) {
@@ -565,7 +566,10 @@ static int tc_aux_link_setup(struct tc_data *tc)
 	value |= SYSCLK_SEL_LSCLK | LSCLK_DIV_2;
 	tc_write(SYS_PLLPARAM, value);
 
-	tc_write(DP_PHY_CTRL, BGREN | PWR_SW_EN | PHY_2LANE | PHY_A0_EN);
+	dp_phy_ctrl = BGREN | PWR_SW_EN | PHY_A0_EN;
+	if (tc->link.base.num_lanes == 2)
+		dp_phy_ctrl |= PHY_2LANE;
+	tc_write(DP_PHY_CTRL, dp_phy_ctrl);
 
 	/*
 	 * Initially PLLs are in bypass. Force PLL parameter update,
@@ -858,7 +862,9 @@ static int tc_main_link_setup(struct tc_data *tc)
 	tc_write(SYS_PLLPARAM, value);
 
 	/* Setup Main Link */
-	dp_phy_ctrl = BGREN | PWR_SW_EN | PHY_2LANE | PHY_A0_EN |  PHY_M0_EN;
+	dp_phy_ctrl = BGREN | PWR_SW_EN | PHY_A0_EN | PHY_M0_EN;
+	if (tc->link.base.num_lanes == 2)
+		dp_phy_ctrl |= PHY_2LANE;
 	tc_write(DP_PHY_CTRL, dp_phy_ctrl);
 	msleep(100);
 
Signature · 6 lines
2.19.1

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
author maintainer
#6 lore raw

PATCH 7/16 drm/bridge: tc358767: reject modes which require too much BW

In reply to Sasha Levin

From: Tomi Valkeinen <[email protected]>

[ Upstream commit 51b9e62eb6950c762162ab7eb8390990179be067 ]

The current driver accepts any videomode with pclk < 154MHz. This is not correct, as with 1 lane and/or 1.62Mbps speed not all videomodes can be supported.

Add code to reject modes that require more bandwidth that is available.

---
 drivers/gpu/drm/bridge/tc358767.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
drivers/gpu/drm/bridge/tc358767.c +10 -0
index 24bb6bfa36f3..792a2548d3bb 100644
@@ -1112,10 +1112,20 @@ static bool tc_bridge_mode_fixup(struct drm_bridge *bridge,
 static int tc_connector_mode_valid(struct drm_connector *connector,
 				   struct drm_display_mode *mode)
 {
+	struct tc_data *tc = connector_to_tc(connector);
+	u32 req, avail;
+	u32 bits_per_pixel = 24;
+
 	/* DPI interface clock limitation: upto 154 MHz */
 	if (mode->clock > 154000)
 		return MODE_CLOCK_HIGH;
 
+	req = mode->clock * bits_per_pixel / 8;
+	avail = tc->link.base.num_lanes * tc->link.base.rate;
+
+	if (req > avail)
+		return MODE_BAD;
+
 	return MODE_OK;
 }
 
Signature · 6 lines
2.19.1

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
author maintainer
#7 lore raw

PATCH 8/16 drm/bridge: tc358767: fix output H/V syncs

In reply to Sasha Levin

From: Tomi Valkeinen <[email protected]>

[ Upstream commit 7923e09c7a766e2d58de7fc395bb84c18e5bc625 ]

The H and V syncs of the DP output are always set to active high. This patch fixes the syncs by configuring them according to the videomode.

---
 drivers/gpu/drm/bridge/tc358767.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/gpu/drm/bridge/tc358767.c +5 -1
index 792a2548d3bb..6eebd8ad0c52 100644
@@ -96,6 +96,8 @@
 #define DP0_STARTVAL		0x064c
 #define DP0_ACTIVEVAL		0x0650
 #define DP0_SYNCVAL		0x0654
+#define SYNCVAL_HS_POL_ACTIVE_LOW	(1 << 15)
+#define SYNCVAL_VS_POL_ACTIVE_LOW	(1 << 31)
 #define DP0_MISC		0x0658
 #define TU_SIZE_RECOMMENDED		(63) /* LSCLK cycles per TU */
 #define BPC_6				(0 << 5)
@@ -724,7 +726,9 @@ static int tc_set_video_mode(struct tc_data *tc, struct drm_display_mode *mode)
 
 	tc_write(DP0_ACTIVEVAL, (mode->vdisplay << 16) | (mode->hdisplay));
 
-	tc_write(DP0_SYNCVAL, (vsync_len << 16) | (hsync_len << 0));
+	tc_write(DP0_SYNCVAL, (vsync_len << 16) | (hsync_len << 0) |
+		 ((mode->flags & DRM_MODE_FLAG_NHSYNC) ? SYNCVAL_HS_POL_ACTIVE_LOW : 0) |
+		 ((mode->flags & DRM_MODE_FLAG_NVSYNC) ? SYNCVAL_VS_POL_ACTIVE_LOW : 0));
 
 	tc_write(DPIPXLFMT, VS_POL_ACTIVE_LOW | HS_POL_ACTIVE_LOW |
 		 DE_POL_ACTIVE_HIGH | SUB_CFG_TYPE_CONFIG1 | DPI_BPP_RGB888);
Signature · 6 lines
2.19.1

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
author maintainer
#12 lore raw

PATCH 13/16 gpio: pl061: handle failed allocations

In reply to Sasha Levin

From: Nicholas Mc Guire <[email protected]>

[ Upstream commit df209c43a0e8258e096fb722dfbdae4f0dd13fde ]

devm_kzalloc(), devm_kstrdup() and devm_kasprintf() all can fail internal allocation and return NULL. Using any of the assigned objects without checking is not safe. As this is early in the boot phase and these allocations really should not fail, any failure here is probably an indication of a more serious issue so it makes little sense to try and rollback the previous allocated resources or try to continue; but rather the probe function is simply exited with -ENOMEM.

Fixes 684284b64aae ("ARM: integrator: add MMCI device to IM-PD1")
---
 arch/arm/mach-integrator/impd1.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
arch/arm/mach-integrator/impd1.c +5 -1
index a109f6482413..0f916c245a2e 100644
@@ -393,7 +393,11 @@ static int __ref impd1_probe(struct lm_device *dev)
 					      sizeof(*lookup) + 3 * sizeof(struct gpiod_lookup),
 					      GFP_KERNEL);
 			chipname = devm_kstrdup(&dev->dev, devname, GFP_KERNEL);
-			mmciname = kasprintf(GFP_KERNEL, "lm%x:00700", dev->id);
+			mmciname = devm_kasprintf(&dev->dev, GFP_KERNEL,
+						  "lm%x:00700", dev->id);
+			if (!lookup || !chipname || !mmciname)
+				return -ENOMEM;
+
 			lookup->dev_id = mmciname;
 			/*
 			 * Offsets on GPIO block 1:
Signature · 1 line
2.19.1
author maintainer
#1 lore raw

PATCH 2/16 perf test shell: Use a fallback to get the pathname in vfs_getname

In reply to Sasha Levin

From: Arnaldo Carvalho de Melo <[email protected]>

[ Upstream commit 03fa483821c0b4db7c2b1453d3332f397d82313f ]

Some kernels, like 4.19.13-300.fc29.x86_64 in fedora 29, fail with the existing probe definition asking for the contents of result->name, working when we ask for the 'filename' variable instead, so add a fallback to that.

Now those tests are back working on fedora 29 systems with that kernel:

  # perf test vfs_getname
  65: Use vfs_getname probe to get syscall args filenames   : Ok
  66: Add vfs_getname probe to get syscall args filenames   : Ok
  67: Check open filename arg using perf trace + vfs_getname: Ok
  #
---
 tools/perf/tests/shell/lib/probe_vfs_getname.sh | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
tools/perf/tests/shell/lib/probe_vfs_getname.sh +2 -1
index 30a950c9d407..068d463e5cbf 100644
@@ -13,7 +13,8 @@ add_probe_vfs_getname() {
 	local verbose=$1
 	if [ $had_vfs_getname -eq 1 ] ; then
 		line=$(perf probe -L getname_flags 2>&1 | egrep 'result.*=.*filename;' | sed -r 's/[[:space:]]+([[:digit:]]+)[[:space:]]+result->uptr.*/\1/')
-		perf probe $verbose "vfs_getname=getname_flags:${line} pathname=result->name:string"
+		perf probe -q       "vfs_getname=getname_flags:${line} pathname=result->name:string" || \
+		perf probe $verbose "vfs_getname=getname_flags:${line} pathname=filename:string"
 	fi
 }
 
Signature · 1 line
2.19.1