dri-devel July 2018

dri-devel@lists.freedesktop.org

277 participants
1883 discussions

[PATCH] fbdev: fbmem: mark expected switch fall-through
by Gustavo A. R. Silva 24 Jul '18

24 Jul '18

In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. Signed-off-by: Gustavo A. R. Silva <gustavo(a)embeddedor.com> --- drivers/video/fbdev/core/fbmem.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c index 609438d..5f3133b 100644 --- a/drivers/video/fbdev/core/fbmem.c +++ b/drivers/video/fbdev/core/fbmem.c @@ -1347,6 +1347,7 @@ static long fb_compat_ioctl(struct file *file, unsigned int cmd, case FBIOGET_CON2FBMAP: case FBIOPUT_CON2FBMAP: arg = (unsigned long) compat_ptr(arg); + /* fall through */ case FBIOBLANK: ret = do_fb_ioctl(info, cmd, arg); break; -- 2.7.4

2 1

[PATCH] fbdev/via: fix defined but not used warning
by Randy Dunlap 24 Jul '18

24 Jul '18

From: Randy Dunlap <rdunlap(a)infradead.org> Fix a build warning in viafbdev.c when CONFIG_PROC_FS is not enabled by marking the unused function as __maybe_unused. ../drivers/video/fbdev/via/viafbdev.c:1471:12: warning: 'viafb_sup_odev_proc_show' defined but not used [-Wunused-function] Signed-off-by: Randy Dunlap <rdunlap(a)infradead.org> Cc: Florian Tobias Schandinat <FlorianSchandinat(a)gmx.de> Cc: linux-fbdev(a)vger.kernel.org Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie(a)samsung.com> Cc: dri-devel(a)lists.freedesktop.org --- drivers/video/fbdev/via/viafbdev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) --- linux-next-20180706.orig/drivers/video/fbdev/via/viafbdev.c +++ linux-next-20180706/drivers/video/fbdev/via/viafbdev.c @@ -19,6 +19,7 @@ * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ +#include <linux/compiler.h> #include <linux/module.h> #include <linux/seq_file.h> #include <linux/slab.h> @@ -1468,7 +1469,7 @@ static const struct file_operations viaf #endif /* CONFIG_FB_VIA_DIRECT_PROCFS */ -static int viafb_sup_odev_proc_show(struct seq_file *m, void *v) +static int __maybe_unused viafb_sup_odev_proc_show(struct seq_file *m, void *v) { via_odev_to_seq(m, supported_odev_map[ viaparinfo->shared->chip_info.gfx_chip_name]);

2 1

[PATCH] omapfb: rename omap2 module to omap2fb.ko
by Arnd Bergmann 24 Jul '18

24 Jul '18

In a kernel configuration with both CONFIG_FB_OMAP=m and CONFIG_FB_OMAP2=m, Kbuild fails to point out that we have two modules with the same name (omapfb.ko), but instead fails with a cryptic error message like: ERROR: "omapfb_register_panel" [drivers/video/fbdev/omap/lcd_osk.ko] undefined! This can now happen when building a randconfig kernel with CONFIG_ARCH_OMAP1, as the omap1 fbdev driver depends on that, whiel the omap2 fbdev driver can now be built anywhere with CONFIG_COMPILE_TEST. The solution is to rename one of the two modules, so for consistency with the directory naming I decided to rename the omap2 version to omap2fb.ko. Fixes: 7378f1149884 ("media: omap2: omapfb: allow building it with COMPILE_TEST") Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- drivers/video/fbdev/omap2/omapfb/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/video/fbdev/omap2/omapfb/Makefile b/drivers/video/fbdev/omap2/omapfb/Makefile index 602edfed09df..f54c3f56b641 100644 --- a/drivers/video/fbdev/omap2/omapfb/Makefile +++ b/drivers/video/fbdev/omap2/omapfb/Makefile @@ -2,5 +2,5 @@ obj-$(CONFIG_OMAP2_VRFB) += vrfb.o obj-y += dss/ obj-y += displays/ -obj-$(CONFIG_FB_OMAP2) += omapfb.o -omapfb-y := omapfb-main.o omapfb-sysfs.o omapfb-ioctl.o +obj-$(CONFIG_FB_OMAP2) += omap2fb.o +omap2fb-y := omapfb-main.o omapfb-sysfs.o omapfb-ioctl.o -- 2.9.0

3 2

[PATCH] video: goldfishfb: fix memory leak on driver remove
by Anton Vasilyev 24 Jul '18

24 Jul '18

goldfish_fb_probe() allocates memory for fb, but goldfish_fb_remove() does not have deallocation of fb, which leads to memory leak on probe/remove. The patch adds deallocation into goldfish_fb_remove(). Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Anton Vasilyev <vasilyev(a)ispras.ru> --- drivers/video/fbdev/goldfishfb.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/video/fbdev/goldfishfb.c b/drivers/video/fbdev/goldfishfb.c index 3b70044773b6..9fe7edf725c6 100644 --- a/drivers/video/fbdev/goldfishfb.c +++ b/drivers/video/fbdev/goldfishfb.c @@ -301,6 +301,7 @@ static int goldfish_fb_remove(struct platform_device *pdev) dma_free_coherent(&pdev->dev, framesize, (void *)fb->fb.screen_base, fb->fb.fix.smem_start); iounmap(fb->reg_base); + kfree(fb); return 0; } -- 2.18.0

2 1

[PATCH] fbdev: fix modedb docs in fb_find_mode
by Daniel Vetter 24 Jul '18

24 Jul '18

Fix up the indenting that confused sphinx. To make sure we don't have to make the examples unreadable with escaping just put them in as block quotes, that seems the simplest solution. Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie(a)samsung.com> Cc: linux-fbdev(a)vger.kernel.org Signed-off-by: Daniel Vetter <daniel.vetter(a)ffwll.ch> --- drivers/video/fbdev/core/modedb.c | 62 ++++++++++++++++--------------- 1 file changed, 32 insertions(+), 30 deletions(-) diff --git a/drivers/video/fbdev/core/modedb.c b/drivers/video/fbdev/core/modedb.c index 2510fa728d77..5ab8af271286 100644 --- a/drivers/video/fbdev/core/modedb.c +++ b/drivers/video/fbdev/core/modedb.c @@ -628,45 +628,47 @@ static int fb_try_mode(struct fb_var_screeninfo *var, struct fb_info *info, } /** - * fb_find_mode - finds a valid video mode - * @var: frame buffer user defined part of display - * @info: frame buffer info structure - * @mode_option: string video mode to find - * @db: video mode database - * @dbsize: size of @db - * @default_mode: default video mode to fall back to - * @default_bpp: default color depth in bits per pixel + * fb_find_mode - finds a valid video mode + * @var: frame buffer user defined part of display + * @info: frame buffer info structure + * @mode_option: string video mode to find + * @db: video mode database + * @dbsize: size of @db + * @default_mode: default video mode to fall back to + * @default_bpp: default color depth in bits per pixel * - * Finds a suitable video mode, starting with the specified mode - * in @mode_option with fallback to @default_mode. If - * @default_mode fails, all modes in the video mode database will - * be tried. + * Finds a suitable video mode, starting with the specified mode + * in @mode_option with fallback to @default_mode. If + * @default_mode fails, all modes in the video mode database will + * be tried. * - * Valid mode specifiers for @mode_option: + * Valid mode specifiers for @mode_option:: * - * <xres>x<yres>[M][R][-<bpp>][@<refresh>][i][m] or - * <name>[-<bpp>][@<refresh>] + * <xres>x<yres>[M][R][-<bpp>][@<refresh>][i][m] + * + * or :: * - * with <xres>, <yres>, <bpp> and <refresh> decimal numbers and - * <name> a string. + * <name>[-<bpp>][@<refresh>] * - * If 'M' is present after yres (and before refresh/bpp if present), - * the function will compute the timings using VESA(tm) Coordinated - * Video Timings (CVT). If 'R' is present after 'M', will compute with - * reduced blanking (for flatpanels). If 'i' is present, compute - * interlaced mode. If 'm' is present, add margins equal to 1.8% - * of xres rounded down to 8 pixels, and 1.8% of yres. The char - * 'i' and 'm' must be after 'M' and 'R'. Example: + * with <xres>, <yres>, <bpp> and <refresh> decimal numbers and + * <name> a string. * - * 1024x768MR-8@60m - Reduced blank with margins at 60Hz. + * If 'M' is present after yres (and before refresh/bpp if present), + * the function will compute the timings using VESA(tm) Coordinated + * Video Timings (CVT). If 'R' is present after 'M', will compute with + * reduced blanking (for flatpanels). If 'i' is present, compute + * interlaced mode. If 'm' is present, add margins equal to 1.8% + * of xres rounded down to 8 pixels, and 1.8% of yres. The char + * 'i' and 'm' must be after 'M' and 'R'. Example:: * - * NOTE: The passed struct @var is _not_ cleared! This allows you - * to supply values for e.g. the grayscale and accel_flags fields. + * 1024x768MR-8@60m - Reduced blank with margins at 60Hz. * - * Returns zero for failure, 1 if using specified @mode_option, - * 2 if using specified @mode_option with an ignored refresh rate, - * 3 if default mode is used, 4 if fall back to any valid mode. + * NOTE: The passed struct @var is _not_ cleared! This allows you + * to supply values for e.g. the grayscale and accel_flags fields. * + * Returns zero for failure, 1 if using specified @mode_option, + * 2 if using specified @mode_option with an ignored refresh rate, + * 3 if default mode is used, 4 if fall back to any valid mode. */ int fb_find_mode(struct fb_var_screeninfo *var, -- 2.18.0

2 1

[PATCH] drm/msm: mark PM functions as __maybe_unused
by Arnd Bergmann 24 Jul '18

24 Jul '18

The suspend/resume functions are not referenced when power management is disabled: drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c:1288:12: error: 'dpu_runtime_resume' defined but not used [-Werror=unused-function] drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c:1261:12: error: 'dpu_runtime_suspend' defined but not used [-Werror=unused-function] This marks them as __maybe_unused to let the compiler drop the functions without complaining. Fixes: 591225291ca2 ("drm/msm: Add SDM845 DPU support") Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c index 8d4678d29cc7..1c0838801e78 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c @@ -1258,7 +1258,7 @@ static int dpu_dev_remove(struct platform_device *pdev) return 0; } -static int dpu_runtime_suspend(struct device *dev) +static int __maybe_unused dpu_runtime_suspend(struct device *dev) { int rc = -1; struct platform_device *pdev = to_platform_device(dev); @@ -1285,7 +1285,7 @@ static int dpu_runtime_suspend(struct device *dev) return rc; } -static int dpu_runtime_resume(struct device *dev) +static int __maybe_unused dpu_runtime_resume(struct device *dev) { int rc = -1; struct platform_device *pdev = to_platform_device(dev); -- 2.18.0

1 0

[PATCH v2 1/3] drm: writeback: Fix doc that says connector should be disconnected
by Alexandru Gheorghe 24 Jul '18

24 Jul '18

During iteration process one of the proposed mechanism for not breaking existing userspace was to report writeback connectors as disconnected, however the final version used DRM_CLIENT_CAP_WRITEBACK_CONNECTORS for that purpose. Signed-off-by: Alexandru Gheorghe <alexandru-cosmin.gheorghe(a)arm.com> --- drivers/gpu/drm/drm_writeback.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/drm_writeback.c b/drivers/gpu/drm/drm_writeback.c index 8273950..e7b6e5e 100644 --- a/drivers/gpu/drm/drm_writeback.c +++ b/drivers/gpu/drm/drm_writeback.c @@ -23,8 +23,8 @@ * from a CRTC to a memory buffer. They are used and act similarly to other * types of connectors, with some important differences: * - Writeback connectors don't provide a way to output visually to the user. - * - Writeback connectors should always report as "disconnected" (so that - * clients which don't understand them will ignore them). + * - Writeback connectors are visible to userspace only when the client sets + * DRM_CLIENT_CAP_WRITEBACK_CONNECTORS. * - Writeback connectors don't have EDID. * * A framebuffer may only be attached to a writeback connector when the -- 2.7.4

11 39

[PATCH] video: fbdev: metronomefb: fix some off by one bugs
by Dan Carpenter 24 Jul '18

24 Jul '18

The "mem" buffer has "size" bytes. The ">" should be ">=" to prevent reading one character beyond the end of the array. Signed-off-by: Dan Carpenter <dan.carpenter(a)oracle.com> --- Not tested. diff --git a/drivers/video/fbdev/metronomefb.c b/drivers/video/fbdev/metronomefb.c index 9085e9525341..bb4fee52e501 100644 --- a/drivers/video/fbdev/metronomefb.c +++ b/drivers/video/fbdev/metronomefb.c @@ -233,7 +233,7 @@ static int load_waveform(u8 *mem, size_t size, int m, int t, /* check temperature range table checksum */ cksum_idx = sizeof(*wfm_hdr) + wfm_hdr->trc + 1; - if (cksum_idx > size) + if (cksum_idx >= size) return -EINVAL; cksum = calc_cksum(sizeof(*wfm_hdr), cksum_idx, mem); if (cksum != mem[cksum_idx]) { @@ -245,7 +245,7 @@ static int load_waveform(u8 *mem, size_t size, int m, int t, /* check waveform mode table address checksum */ wmta = get_unaligned_le32(wfm_hdr->wmta) & 0x00FFFFFF; cksum_idx = wmta + m*4 + 3; - if (cksum_idx > size) + if (cksum_idx >= size) return -EINVAL; cksum = calc_cksum(cksum_idx - 3, cksum_idx, mem); if (cksum != mem[cksum_idx]) { @@ -257,7 +257,7 @@ static int load_waveform(u8 *mem, size_t size, int m, int t, /* check waveform temperature table address checksum */ tta = get_unaligned_le32(mem + wmta + m * 4) & 0x00FFFFFF; cksum_idx = tta + trn*4 + 3; - if (cksum_idx > size) + if (cksum_idx >= size) return -EINVAL; cksum = calc_cksum(cksum_idx - 3, cksum_idx, mem); if (cksum != mem[cksum_idx]) { @@ -270,7 +270,7 @@ static int load_waveform(u8 *mem, size_t size, int m, int t, metromem buffer. this does runlength decoding of the waveform */ wfm_idx = get_unaligned_le32(mem + tta + trn * 4) & 0x00FFFFFF; owfm_idx = wfm_idx; - if (wfm_idx > size) + if (wfm_idx >= size) return -EINVAL; while (wfm_idx < size) { unsigned char rl; @@ -292,7 +292,7 @@ static int load_waveform(u8 *mem, size_t size, int m, int t, } cksum_idx = wfm_idx; - if (cksum_idx > size) + if (cksum_idx >= size) return -EINVAL; cksum = calc_cksum(owfm_idx, cksum_idx, mem); if (cksum != mem[cksum_idx]) {

2 1

[PATCH] video/console/vgacon: Print big fat warning with nomodeset
by Lyude Paul 24 Jul '18

24 Jul '18

It's been a pretty good while since kernel modesetting was introduced. It has almost entirely replaced previous solutions which required userspace modesetting, and I can't even recall any drivers off the top of my head for modern day hardware that don't only support one or the other. Even nvidia's ugly blob does not require the use of nomodeset, and only requires that nouveau be blacklisted. Effectively, the only thing nomodeset does in the year 2018 is disable your graphics drivers. Since VESA is a thing, this will give many users the false impression that they've actually fixed an issue they were having with their machine simply because the laptop will boot up to a degraded GUI. This of course, is never actually the case. Things get even worse when you consider that there's still an enormous amount of tutorials users find on the internet that still suggest adding nomodeset, along with various users who have been around long enough to still suggest it. There really isn't any legitimate reason I can see for this to be an option that's used by anyone else other then developers, or properly informed users. So, let's end the confusion and start printing warnings whenever it's enabled. Signed-off-by: Lyude Paul <lyude(a)redhat.com> --- drivers/video/console/vgacon.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c index f09e17b60e45..09731b2f6815 100644 --- a/drivers/video/console/vgacon.c +++ b/drivers/video/console/vgacon.c @@ -112,6 +112,11 @@ EXPORT_SYMBOL(vgacon_text_force); static int __init text_mode(char *str) { vgacon_text_mode_force = true; + + pr_warning("You have booted with nomodeset. This means your GPU drivers are DISABLED\n"); + pr_warning("Any video related functionality will be severely degraded, and you may not even be able to suspend the system properly\n"); + pr_warning("Unless you actually understand what nomodeset does, you should reboot without enabling it\n"); + return 1; } -- 2.17.1

4 3

[PATCH v2 fbdev-for-next 2/2] fbdev: fix typo in comment
by Yisheng Xie 24 Jul '18

24 Jul '18

change beeng to being and occured to occurred. Signed-off-by: Yisheng Xie <ysxie(a)foxmail.com> --- v2: - new add one. include/linux/fb.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/include/linux/fb.h b/include/linux/fb.h index fd31e6f..3e7e753 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -126,7 +126,7 @@ struct fb_cursor_user { /* The resolution of the passed in fb_info about to change */ #define FB_EVENT_MODE_CHANGE 0x01 -/* The display on this fb_info is beeing suspended, no access to the +/* The display on this fb_info is being suspended, no access to the * framebuffer is allowed any more after that call returns */ #define FB_EVENT_SUSPEND 0x02 @@ -159,9 +159,9 @@ struct fb_cursor_user { #define FB_EVENT_FB_UNBIND 0x0E /* CONSOLE-SPECIFIC: remap all consoles to new fb - for vga_switcheroo */ #define FB_EVENT_REMAP_ALL_CONSOLE 0x0F -/* A hardware display blank early change occured */ +/* A hardware display blank early change occurred */ #define FB_EARLY_EVENT_BLANK 0x10 -/* A hardware display blank revert early change occured */ +/* A hardware display blank revert early change occurred */ #define FB_R_EARLY_EVENT_BLANK 0x11 struct fb_event { -- 1.9.1

2 1

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel July 2018