similar to: [PATCH] drm/nouveau: add missing single_release()

Reviewed-by: Karol Herbst <kherbst at redhat.com> On Thu, Sep 5, 2019 at 9:14 AM Wei Yongjun <weiyongjun1 at huawei.com> wrote: > > When using single_open() for opening, single_release() should be > used, otherwise there is a memory leak. > > This is detected by Coccinelle semantic patch. > > Fixes: 6e9fc177399f ("drm/nouveau/debugfs: add copy of sysfs pstate

I made a little changes in this series: 1. merge the two last patches together 2. unify the private data interface with the drm debugfs one now it should be very obvious for a new dev on how to add new files to debugfs and how to get access to the nouveau structs Karol Herbst (5): debugfs: add infrastructure to add files with other fops than only read debugfs: rename functions to

Reviewed-by: Lyude Paul <lyude at redhat.com> Thanks! On Tue, 2020-07-21 at 15:17 +0000, Wei Yongjun wrote: > When using single_open() for opening, single_release() should be > used instead of seq_release(), otherwise there is a memory leak. > > Fixes: 12885ecbfe62 ("drm/nouveau/kms/nvd9-: Add CRC support") > Reported-by: Hulk Robot <hulkci at huawei.com>

Signed-off-by: Karol Herbst <nouveau at karolherbst.de> --- drm/nouveau/nouveau_debugfs.c | 76 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 76 insertions(+) diff --git a/drm/nouveau/nouveau_debugfs.c b/drm/nouveau/nouveau_debugfs.c index 3d0dc19..31b309f 100644 --- a/drm/nouveau/nouveau_debugfs.c +++ b/drm/nouveau/nouveau_debugfs.c @@ -180,6 +180,81 @@ static const struct

Since 987d65d01356 (drm: debugfs: make drm_debugfs_create_files() never fail), there is no need to ever check for the the return value of debugfs_create_file() and drm_debugfs_create_files(). Therefore, remove unnecessary checks and error handling in nouveau_drm_debugfs_init. These changes also enable nouveau_drm_debugfs_init() to be declared as void. Signed-off-by: Wambui Karuga

As there is no need to check for the return value of debugfs_create_file and drm_debugfs_create_files, remove unnecessary checks and error handling in nouveau_drm_debugfs_init. Signed-off-by: Wambui Karuga <wambui.karugax at gmail.com> --- drivers/gpu/drm/nouveau/nouveau_debugfs.c | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-) diff --git

This is a small patch series that adds a strap_peek file into our debugfs, and sets the size of the vbios.rom debugfs file so that nvbios can easily be used to parse the vbios even on systems where the normal BIOS retrieval methods (for example, laptops that need ACPI to access the vbios for the nvidia GPU) won't work. This should make it a little easier to collect vbioses. Lyude Paul (2):

This hooks up the DRM helpers for dumping information on the current status of each MST topology from nouveau's perspective to debugfs files, similar to what i915 does (albeit, i915 labels their debugfs node for this as i915_dp_mst_info). Lyude Paul (2): drm/nouveau: Expose nv50 MST structures in disp.h drm/nouveau: Hook up dp_mst_info in debugfs drivers/gpu/drm/nouveau/dispnv50/disp.c

This commit adds a function to dump a DRM GPU VA space and a macro for drivers to register the struct drm_info_list 'gpuvas' entry. Most likely, most drivers might maintain one DRM GPU VA space per struct drm_file, but there might also be drivers not having a fixed relation between DRM GPU VA spaces and a DRM core infrastructure, hence we need the indirection via the driver iterating

Add a /proc node for the VMI sub-arch, which gives information on the VMI ROM detected using /proc/vmi/info and a list of kernel annotations in /proc/vmi/annotations. The timing information is VMware specific, and should probably be put into a separate /proc node (and a separate patch for our internal use). Signed-off-by: Zachary Amsden <zach@vmware.com> Index:

Add a /proc node for the VMI sub-arch, which gives information on the VMI ROM detected using /proc/vmi/info and a list of kernel annotations in /proc/vmi/annotations. The timing information is VMware specific, and should probably be put into a separate /proc node (and a separate patch for our internal use). Signed-off-by: Zachary Amsden <zach@vmware.com> Index:

fixes a compile error Signed-off-by: Karol Herbst <nouveau at karolherbst.de> --- drm/nouveau/nouveau_debugfs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drm/nouveau/nouveau_debugfs.h b/drm/nouveau/nouveau_debugfs.h index 52c7161..b8c03ff 100644 --- a/drm/nouveau/nouveau_debugfs.h +++ b/drm/nouveau/nouveau_debugfs.h @@ -34,13 +34,13 @@

Date: Sun, 16 Apr 2023 11:22:23 +0200 Several update suggestions were taken into account from static source code analysis. Markus Elfring (9): debugfs: Move an expression into a function call parameter in nouveau_debugfs_pstate_set() debugfs: Move a variable assignment behind a null pointer check in nouveau_debugfs_pstate_get() debugfs: Use seq_putc() in

When calling debugfs functions, there is no need to ever check the return value. The function can work or not, but the code logic should never do something different based on this. Cc: Ben Skeggs <bskeggs at redhat.com> Cc: David Airlie <airlied at linux.ie> Cc: Daniel Vetter <daniel at ffwll.ch> Cc: dri-devel at lists.freedesktop.org Cc: nouveau at lists.freedesktop.org

From: Matt Parnell <mparnell at gmail.com> Sometimes we have DEBUG_FS enabled, but don't want output from certain modules. Most modules make this an option, so I figured that Nouveau needed it too. Signed-off-by: Matt Parnell <mparnell at gmail.com> --- drivers/gpu/drm/Kconfig | 11 +++++++++++ drivers/gpu/drm/nouveau/Makefile | 2 +-

On 17/11/17 02:04, Karol Herbst wrote: > Fixes various reclocking related issues on prime systems. Is that the only place that was not covered? Could you check if other places would need this code too? In any case, this patch is (assuming you are calling the right functions to prevent the GPU from sleeping): Signed-off-by: Martin Peres <martin.peres at free.fr> > > Signed-off-by:

nouveau_debugfs_strap_peek() calls pm_runtime_get_sync() that increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> --- drivers/gpu/drm/nouveau/nouveau_debugfs.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.c

First patch fixes a potential vbios parsing bug on Fermi GPUs, but it may also affect Kepler ones. Second patch wakes the GPU up whenever the user tries to reclock to prevent the write to the pstate file from freezing. Karol Herbst (2): bios/vpstate: There are some fermi vbios with no boost or tdp entry debugfs: Wake up GPU before doing any reclocking drm/nouveau/nouveau_debugfs.c

nouveau_debugfs(drm) will never be NULL, because we're taking the value of the potentially null device pointer and adding to it so it isn't 0x0. So, check if drm is NULL instead. Signed-off-by: Lyude Paul <lyude at redhat.com> Cc: Karol Herbst <karolherbst at gmail.com> --- drivers/gpu/drm/nouveau/nouveau_debugfs.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)

On Mon, Oct 26, 2015 at 2:13 PM, Karol Herbst <nouveau at karolherbst.de> wrote: > From: Karol Herbst <git at karolherbst.de> > > --- > drm/nouveau/include/nvif/device.h | 1 + > drm/nouveau/include/nvkm/subdev/pmu.h | 10 ++++++++++ > drm/nouveau/nouveau_debugfs.c | 23 +++++++++++++++++++++++ > drm/nouveau/nvkm/subdev/pmu/base.c | 18