From b6ffbab813cfc535db5c3815b4eb16dd7d97197c Mon Sep 17 00:00:00 2001 From: Oded Gabbay Date: Sun, 7 Dec 2014 22:27:24 +0200 Subject: [PATCH 1/7] amdkfd: Fix accounting of device queues This patch fixes a device QCM bug, where the number of queues were not counted correctly for the operation of update queue. The count was incorrect as there was no regard to the previous state of the queue. Signed-off-by: Oded Gabbay --- .../gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 924e90c072e5..f44d6737b65a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -320,6 +320,7 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q) { int retval; struct mqd_manager *mqd; + bool prev_active = false; BUG_ON(!dqm || !q || !q->mqd); @@ -330,10 +331,18 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q) return -ENOMEM; } - retval = mqd->update_mqd(mqd, q->mqd, &q->properties); if (q->properties.is_active == true) + prev_active = true; + + /* + * + * check active state vs. the previous state + * and modify counter accordingly + */ + retval = mqd->update_mqd(mqd, q->mqd, &q->properties); + if ((q->properties.is_active == true) && (prev_active == false)) dqm->queue_count++; - else + else if ((q->properties.is_active == false) && (prev_active == true)) dqm->queue_count--; if (sched_policy != KFD_SCHED_POLICY_NO_HWS) From 8dfead6c2836621bd878626ae841fc55259b2523 Mon Sep 17 00:00:00 2001 From: Ben Goz Date: Tue, 2 Dec 2014 16:41:08 +0200 Subject: [PATCH 2/7] amdkfd: Fixing topology bug in building sysfs nodes Original code sent always 0 as the index number of the node. This patch fixes this bug by sending a variable which is incremented per node. Signed-off-by: Ben Goz Reviewed-by: Oded Gabbay Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index b11792d7e70e..cca1708fd811 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -921,7 +921,7 @@ static int kfd_build_sysfs_node_tree(void) uint32_t i = 0; list_for_each_entry(dev, &topology_device_list, list) { - ret = kfd_build_sysfs_node_entry(dev, 0); + ret = kfd_build_sysfs_node_entry(dev, i); if (ret < 0) return ret; i++; From 0cb989c0c6e62f235b14e48604f909085494026b Mon Sep 17 00:00:00 2001 From: Oded Gabbay Date: Fri, 5 Dec 2014 10:40:34 +0200 Subject: [PATCH 3/7] amdkfd: Remove duplicate include Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 7d4974b83af7..fe5c543599b0 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -31,7 +31,6 @@ #include #include #include -#include #include #include #include "kfd_priv.h" From 1bacc894c227fad8a727eb99728df708eba57654 Mon Sep 17 00:00:00 2001 From: Oded Gabbay Date: Mon, 22 Dec 2014 11:47:37 +0200 Subject: [PATCH 4/7] drivers: Move iommu/ before gpu/ in Makefile MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit AMD GPU devices are dependent on AMD IOMMU controller functionality to allow the GPU to access a process's virtual memory address space, without the need for pinning the memory. This patch changes the order in the drivers makefile, so iommu/ subsystem is linked before gpu/ subsystem. That way, if the gpu and iommu drivers are compiled inside the kernel image (not as modules), the correct order of device loading is still maintained (iommu module is loaded before gpu module). Signed-off-by: Oded Gabbay Reviewed-by: Christian König --- drivers/Makefile | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/Makefile b/drivers/Makefile index 67d2334dc41e..527a6da8d539 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -50,7 +50,10 @@ obj-$(CONFIG_RESET_CONTROLLER) += reset/ obj-y += tty/ obj-y += char/ -# gpu/ comes after char for AGP vs DRM startup +# iommu/ comes before gpu as gpu are using iommu controllers +obj-$(CONFIG_IOMMU_SUPPORT) += iommu/ + +# gpu/ comes after char for AGP vs DRM startup and after iommu obj-y += gpu/ obj-$(CONFIG_CONNECTOR) += connector/ @@ -141,7 +144,6 @@ obj-y += clk/ obj-$(CONFIG_MAILBOX) += mailbox/ obj-$(CONFIG_HWSPINLOCK) += hwspinlock/ -obj-$(CONFIG_IOMMU_SUPPORT) += iommu/ obj-$(CONFIG_REMOTEPROC) += remoteproc/ obj-$(CONFIG_RPMSG) += rpmsg/ From 611a03d764b151190ba7a388b00be72b23aee2bc Mon Sep 17 00:00:00 2001 From: Oded Gabbay Date: Sun, 21 Dec 2014 15:21:41 +0200 Subject: [PATCH 5/7] drm: Put amdkfd before radeon in drm Makefile MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When amdkfd and radeon are compiled inside the kernel image (not as modules), radeon will load before amdkfd, which will cause a bug when radeon will probe the GPUs. When the two drivers are compiled as modules, amdkfd is loaded after radeon is loaded but before radeon starts probing the GPUs. This is done because radeon loads the amdkfd module through symbol_request function. This patch makes amdkfd load before radeon when they are both compiled inside the kernel image, which makes the behavior similar to the case when they are modules, and prevents the kernel bug. Signed-off-by: Oded Gabbay Reviewed-by: Christian König --- drivers/gpu/drm/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index 66e40398b3d3..e620807418ea 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -37,6 +37,7 @@ obj-$(CONFIG_DRM_MIPI_DSI) += drm_mipi_dsi.o obj-$(CONFIG_DRM_TTM) += ttm/ obj-$(CONFIG_DRM_TDFX) += tdfx/ obj-$(CONFIG_DRM_R128) += r128/ +obj-$(CONFIG_HSA_AMD) += amd/amdkfd/ obj-$(CONFIG_DRM_RADEON)+= radeon/ obj-$(CONFIG_DRM_MGA) += mga/ obj-$(CONFIG_DRM_I810) += i810/ @@ -67,4 +68,3 @@ obj-$(CONFIG_DRM_IMX) += imx/ obj-y += i2c/ obj-y += panel/ obj-y += bridge/ -obj-$(CONFIG_HSA_AMD) += amd/amdkfd/ From 68d0cb49f8db5ddce7cb6cbd7781e232c42c472a Mon Sep 17 00:00:00 2001 From: Sasha Levin Date: Sun, 28 Dec 2014 11:44:37 -0500 Subject: [PATCH 6/7] amdkfd: actually allocate longs for the pasid bitmask Commit "amdkfd: use sizeof(long) granularity for the pasid bitmask" calculated the number of longs it will need, but ended up allocating that number of bytes rather than longs. Fix that silly error and allocate the amount of data really required. Signed-off-by: Sasha Levin Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdkfd/kfd_pasid.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c index 71699ad97d74..4c25ef504f79 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c @@ -32,7 +32,7 @@ int kfd_pasid_init(void) { pasid_limit = max_num_of_processes; - pasid_bitmap = kzalloc(BITS_TO_LONGS(pasid_limit), GFP_KERNEL); + pasid_bitmap = kcalloc(BITS_TO_LONGS(pasid_limit), sizeof(long), GFP_KERNEL); if (!pasid_bitmap) return -ENOMEM; From 38c2adfb00db045a876dd667040abc01b788ad61 Mon Sep 17 00:00:00 2001 From: Oded Gabbay Date: Mon, 22 Dec 2014 11:19:23 +0200 Subject: [PATCH 7/7] drm/radeon: Init amdkfd only if it was compiled MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch changes the radeon_kfd_init(), which is used to initialize the interface between radeon and amdkfd, so the interface will be initialized only if amdkfd was build, either as module or inside the kernel image. In the modules case, the symbol_request() will be used (same as old code). In the in-image compilation case, a direct call to kgd2kfd_init() will be done. For other cases, radeon_kfd_init() will just return false. This patch is necessary because in case of the following specific configuration: kernel 32-bit, no modules support, random kernel base and no hibernation, the symbol_request() doesn't work as expected - it doesn't return NULL if the symbol doesn't exists - which makes the kernel panic. Reviewed-by: Christian König Signed-off-by: Oded Gabbay --- drivers/gpu/drm/radeon/radeon_kfd.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/gpu/drm/radeon/radeon_kfd.c b/drivers/gpu/drm/radeon/radeon_kfd.c index 242fd8b1b221..d3e78b4a6b75 100644 --- a/drivers/gpu/drm/radeon/radeon_kfd.c +++ b/drivers/gpu/drm/radeon/radeon_kfd.c @@ -101,6 +101,7 @@ static const struct kgd2kfd_calls *kgd2kfd; bool radeon_kfd_init(void) { +#if defined(CONFIG_HSA_AMD_MODULE) bool (*kgd2kfd_init_p)(unsigned, const struct kfd2kgd_calls*, const struct kgd2kfd_calls**); @@ -117,6 +118,17 @@ bool radeon_kfd_init(void) } return true; +#elif defined(CONFIG_HSA_AMD) + if (!kgd2kfd_init(KFD_INTERFACE_VERSION, &kfd2kgd, &kgd2kfd)) { + kgd2kfd = NULL; + + return false; + } + + return true; +#else + return false; +#endif } void radeon_kfd_fini(void)