Hello,
syzbot found the following crash on:
HEAD commit: 262f7a6b Merge tag 'for-5.7-rc3-tag' of git://git.kernel.o.. git tree: upstream console output: https://syzkaller.appspot.com/x/log.txt?x=12786888100000 kernel config: https://syzkaller.appspot.com/x/.config?x=5b075813ec8b93cd dashboard link: https://syzkaller.appspot.com/bug?extid=1f29e126cf461c4de3b3 compiler: gcc (GCC) 9.0.0 20181231 (experimental)
Unfortunately, I don't have any reproducer for this crash yet.
IMPORTANT: if you fix the bug, please add the following tag to the commit: Reported-by: syzbot+1f29e126cf461c4de3b3@syzkaller.appspotmail.com
BUG: unable to handle page fault for address: ffff8880ffca0e80 #PF: supervisor write access in kernel mode #PF: error_code(0x0002) - not-present page PGD d401067 P4D d401067 PUD 0 Oops: 0002 [#1] PREEMPT SMP KASAN CPU: 0 PID: 710 Comm: syz-executor.5 Not tainted 5.7.0-rc3-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:writeb arch/x86/include/asm/io.h:65 [inline] RIP: 0010:vga_imageblit_expand drivers/video/fbdev/vga16fb.c:1168 [inline] RIP: 0010:vga16fb_imageblit+0xa5b/0x2210 drivers/video/fbdev/vga16fb.c:1260 Code: 48 89 fa 48 c1 ea 03 0f b6 14 02 48 89 f8 83 e0 07 83 c0 03 38 d0 7c 09 84 d2 74 05 e8 ee 59 ed fd 41 8b 47 14 48 8b 74 24 08 <88> 06 0f ae e8 8a 06 b8 05 00 00 00 ba ce 03 00 00 ee 48 c7 c2 18 RSP: 0000:ffffc90002ea71f0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000007 RCX: ffffc90014391000 RDX: 0000000000000000 RSI: ffff8880ffca0e80 RDI: ffffc90002ea739c RBP: ffffc90002ea738c R08: ffff8880922ac200 R09: 0000000000000000 R10: ffffffff8a895007 R11: fffffbfff1512a00 R12: 0000000000000000 R13: ffff888218de5140 R14: 0000000000000001 R15: ffffc90002ea7388 FS: 00007fbeeb282700(0000) GS:ffff8880ae600000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffff8880ffca0e80 CR3: 000000008e9c5000 CR4: 00000000001406f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: bit_putcs_unaligned drivers/video/fbdev/core/bitblit.c:139 [inline] bit_putcs+0x910/0xe10 drivers/video/fbdev/core/bitblit.c:188 fbcon_putcs+0x345/0x3f0 drivers/video/fbdev/core/fbcon.c:1362 con_flush drivers/tty/vt/vt.c:2569 [inline] do_con_write.part.0+0x7d1/0x1dc0 drivers/tty/vt/vt.c:2772 do_con_write drivers/tty/vt/vt.c:2588 [inline] con_write+0x41/0xe0 drivers/tty/vt/vt.c:3154 process_output_block drivers/tty/n_tty.c:595 [inline] n_tty_write+0x3f0/0xf90 drivers/tty/n_tty.c:2333 do_tty_write drivers/tty/tty_io.c:962 [inline] tty_write+0x495/0x800 drivers/tty/tty_io.c:1046 __vfs_write+0x76/0x100 fs/read_write.c:495 __kernel_write+0x11c/0x3a0 fs/read_write.c:516 write_pipe_buf+0x153/0x1e0 fs/splice.c:809 splice_from_pipe_feed fs/splice.c:512 [inline] __splice_from_pipe+0x3e6/0x7b0 fs/splice.c:636 splice_from_pipe+0xd9/0x140 fs/splice.c:671 default_file_splice_write+0x37/0x90 fs/splice.c:821 do_splice_from fs/splice.c:863 [inline] direct_splice_actor+0x115/0x160 fs/splice.c:1037 splice_direct_to_actor+0x38c/0x980 fs/splice.c:992 do_splice_direct+0x1b4/0x280 fs/splice.c:1080 do_sendfile+0x555/0xc50 fs/read_write.c:1521 __do_sys_sendfile64 fs/read_write.c:1582 [inline] __se_sys_sendfile64 fs/read_write.c:1568 [inline] __x64_sys_sendfile64+0x1cc/0x210 fs/read_write.c:1568 do_syscall_64+0xf6/0x7d0 arch/x86/entry/common.c:295 entry_SYSCALL_64_after_hwframe+0x49/0xb3 RIP: 0033:0x45c829 Code: 0d b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 db b6 fb ff c3 66 2e 0f 1f 84 00 00 00 00 RSP: 002b:00007fbeeb281c78 EFLAGS: 00000246 ORIG_RAX: 0000000000000028 RAX: ffffffffffffffda RBX: 00000000004fc0c0 RCX: 000000000045c829 RDX: 0000000000000000 RSI: 0000000000000004 RDI: 0000000000000003 RBP: 000000000078bf00 R08: 0000000000000000 R09: 0000000000000000 R10: 0800000080004103 R11: 0000000000000246 R12: 00000000ffffffff R13: 00000000000008d6 R14: 00000000004cb7a1 R15: 00007fbeeb2826d4 Modules linked in: CR2: ffff8880ffca0e80 ---[ end trace 5bb103c4fc7bf525 ]--- RIP: 0010:writeb arch/x86/include/asm/io.h:65 [inline] RIP: 0010:vga_imageblit_expand drivers/video/fbdev/vga16fb.c:1168 [inline] RIP: 0010:vga16fb_imageblit+0xa5b/0x2210 drivers/video/fbdev/vga16fb.c:1260 Code: 48 89 fa 48 c1 ea 03 0f b6 14 02 48 89 f8 83 e0 07 83 c0 03 38 d0 7c 09 84 d2 74 05 e8 ee 59 ed fd 41 8b 47 14 48 8b 74 24 08 <88> 06 0f ae e8 8a 06 b8 05 00 00 00 ba ce 03 00 00 ee 48 c7 c2 18 RSP: 0000:ffffc90002ea71f0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000007 RCX: ffffc90014391000 RDX: 0000000000000000 RSI: ffff8880ffca0e80 RDI: ffffc90002ea739c RBP: ffffc90002ea738c R08: ffff8880922ac200 R09: 0000000000000000 R10: ffffffff8a895007 R11: fffffbfff1512a00 R12: 0000000000000000 R13: ffff888218de5140 R14: 0000000000000001 R15: ffffc90002ea7388 FS: 00007fbeeb282700(0000) GS:ffff8880ae600000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffff8880ffca0e80 CR3: 000000008e9c5000 CR4: 00000000001406f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
--- This bug is generated by a bot. It may contain errors. See https://goo.gl/tpsmEJ for more information about syzbot. syzbot engineers can be reached at syzkaller@googlegroups.com.
syzbot will keep track of this bug report. See: https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
syzbot has found a reproducer for the following issue on:
HEAD commit: d2b6f8a1 Merge tag 'xfs-5.13-merge-3' of git://git.kernel... git tree: upstream console output: https://syzkaller.appspot.com/x/log.txt?x=11d80be1d00000 kernel config: https://syzkaller.appspot.com/x/.config?x=53fdf14defd48c56 dashboard link: https://syzkaller.appspot.com/bug?extid=1f29e126cf461c4de3b3 compiler: Debian clang version 11.0.1-2 syz repro: https://syzkaller.appspot.com/x/repro.syz?x=16d9ff43d00000 C reproducer: https://syzkaller.appspot.com/x/repro.c?x=10981693d00000
IMPORTANT: if you fix the issue, please add the following tag to the commit: Reported-by: syzbot+1f29e126cf461c4de3b3@syzkaller.appspotmail.com
BUG: unable to handle page fault for address: ffff888001000040 #PF: supervisor write access in kernel mode #PF: error_code(0x0003) - permissions violation PGD 11201067 P4D 11201067 PUD 11202067 PMD 80000000010001e1 Oops: 0003 [#1] PREEMPT SMP KASAN CPU: 0 PID: 8403 Comm: syz-executor112 Not tainted 5.12.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:writeb arch/x86/include/asm/io.h:65 [inline] RIP: 0010:vga_imageblit_expand drivers/video/fbdev/vga16fb.c:1176 [inline] RIP: 0010:vga16fb_imageblit+0xcee/0x1cb0 drivers/video/fbdev/vga16fb.c:1260 Code: 66 66 2e 0f 1f 84 00 00 00 00 00 90 4c 89 e0 48 c1 e8 03 48 b9 00 00 00 00 00 fc ff df 0f b6 04 08 84 c0 75 1b 41 0f b6 04 24 <41> 88 06 85 ed 74 2b 49 ff c4 49 ff c6 e8 80 ae 43 fd ff cd eb cc RSP: 0018:ffffc9000163f0a0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff888001000040 RCX: dffffc0000000000 RDX: ffff888022ad54c0 RSI: 0000000000000002 RDI: 0000000000000000 RBP: 0000000000000001 R08: ffffffff843b289b R09: 0000000000000000 R10: 0000000000000002 R11: ffff888022ad54c0 R12: ffff8880181bcea8 R13: ffffc9000163f2cc R14: ffff888001000040 R15: 0000000000000004 FS: 0000000001207300(0000) GS:ffff8880b9a00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffff888001000040 CR3: 0000000028d32000 CR4: 00000000001506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: bit_putcs+0x18e8/0x1de0 drivers/video/fbdev/core/bitblit.c:105 fbcon_putcs+0x2ae/0x430 drivers/video/fbdev/core/fbcon.c:1296 do_update_region+0x4d6/0x6a0 drivers/tty/vt/vt.c:676 redraw_screen+0x8f6/0x1270 drivers/tty/vt/vt.c:1035 fbcon_blank+0x556/0xa50 drivers/video/fbdev/core/fbcon.c:2207 do_unblank_screen+0x27e/0xb20 drivers/tty/vt/vt.c:4406 vt_kdsetmode drivers/tty/vt/vt_ioctl.c:276 [inline] vt_k_ioctl drivers/tty/vt/vt_ioctl.c:381 [inline] vt_ioctl+0x2a82/0x3180 drivers/tty/vt/vt_ioctl.c:713 tty_ioctl+0xf51/0x1720 drivers/tty/tty_io.c:2805 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:1069 [inline] __se_sys_ioctl+0xfb/0x170 fs/ioctl.c:1055 do_syscall_64+0x3f/0xb0 arch/x86/entry/common.c:47 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x43fef9 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 b1 14 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 c0 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007ffc931a4c48 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 0000000000013f84 RCX: 000000000043fef9 RDX: 0000000000000000 RSI: 0000000000004b3a RDI: 0000000000000003 RBP: 0000000000000000 R08: 000000000000000d R09: 00007ffc931a4de8 R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffc931a4c5c R13: 431bde82d7b634db R14: 00000000004ae018 R15: 0000000000400488 Modules linked in: CR2: ffff888001000040 ---[ end trace 96734cf7ef5cce91 ]--- RIP: 0010:writeb arch/x86/include/asm/io.h:65 [inline] RIP: 0010:vga_imageblit_expand drivers/video/fbdev/vga16fb.c:1176 [inline] RIP: 0010:vga16fb_imageblit+0xcee/0x1cb0 drivers/video/fbdev/vga16fb.c:1260 Code: 66 66 2e 0f 1f 84 00 00 00 00 00 90 4c 89 e0 48 c1 e8 03 48 b9 00 00 00 00 00 fc ff df 0f b6 04 08 84 c0 75 1b 41 0f b6 04 24 <41> 88 06 85 ed 74 2b 49 ff c4 49 ff c6 e8 80 ae 43 fd ff cd eb cc RSP: 0018:ffffc9000163f0a0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff888001000040 RCX: dffffc0000000000 RDX: ffff888022ad54c0 RSI: 0000000000000002 RDI: 0000000000000000 RBP: 0000000000000001 R08: ffffffff843b289b R09: 0000000000000000 R10: 0000000000000002 R11: ffff888022ad54c0 R12: ffff8880181bcea8 R13: ffffc9000163f2cc R14: ffff888001000040 R15: 0000000000000004 FS: 0000000001207300(0000) GS:ffff8880b9a00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffff888001000040 CR3: 0000000028d32000 CR4: 00000000001506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
syzbot has bisected this issue to:
commit 988d0763361bb65690d60e2bc53a6b72777040c3 Author: Tetsuo Handa penguin-kernel@i-love.sakura.ne.jp Date: Sun Sep 27 11:46:30 2020 +0000
vt_ioctl: make VT_RESIZEX behave like VT_RESIZE
bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=15633759d00000 start commit: d2b6f8a1 Merge tag 'xfs-5.13-merge-3' of git://git.kernel... git tree: upstream final oops: https://syzkaller.appspot.com/x/report.txt?x=17633759d00000 console output: https://syzkaller.appspot.com/x/log.txt?x=13633759d00000 kernel config: https://syzkaller.appspot.com/x/.config?x=53fdf14defd48c56 dashboard link: https://syzkaller.appspot.com/bug?extid=1f29e126cf461c4de3b3 syz repro: https://syzkaller.appspot.com/x/repro.syz?x=16d9ff43d00000 C reproducer: https://syzkaller.appspot.com/x/repro.c?x=10981693d00000
Reported-by: syzbot+1f29e126cf461c4de3b3@syzkaller.appspotmail.com Fixes: 988d0763361b ("vt_ioctl: make VT_RESIZEX behave like VT_RESIZE")
For information about bisection process see: https://goo.gl/tpsmEJ#bisection
On 2021/05/02 10:53, syzbot wrote:
syzbot has bisected this issue to:
commit 988d0763361bb65690d60e2bc53a6b72777040c3 Author: Tetsuo Handa penguin-kernel@i-love.sakura.ne.jp Date: Sun Sep 27 11:46:30 2020 +0000
vt_ioctl: make VT_RESIZEX behave like VT_RESIZE
That commit is irrelevant. Below is a simplified reproducer.
---------- #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <sys/ioctl.h> #include <linux/kd.h> #include <linux/vt.h>
int main(int argc, char *argv[]) { const int fd = open("/dev/char/4:1", O_RDWR); struct vt_sizes vt = { 0x4100, 2 };
ioctl(fd, KDSETMODE, KD_GRAPHICS); ioctl(fd, VT_RESIZE, &vt); ioctl(fd, KDSETMODE, KD_TEXT); return 0; } ----------
In vga16fb_probe(), we have
---------- /* XXX share VGA_FB_PHYS and I/O region with vgacon and others */ info->screen_base = (void __iomem *)VGA_MAP_MEM(VGA_FB_PHYS, 0);
if (!info->screen_base) { printk(KERN_ERR "vga16fb: unable to map device\n"); ret = -ENOMEM; goto err_ioremap; }
printk(KERN_INFO "vga16fb: mapped to 0x%p\n", info->screen_base); ----------
and ffff8880000a0000 is assigned for 80 x 30 screen upon boot.
---------- [ 4.584361][ T1] vga16fb: mapped to 0xffff8880000a0000 [ 6.137556][ T1] Console: switching to colour frame buffer device 80x30 [ 7.829276][ T1] fb0: VGA16 VGA frame buffer device ----------
With debug printk() patch shown below,
---------- diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 01645e87b3d5..af860b12db44 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -641,6 +641,8 @@ static void con_scroll(struct vc_data *vc, unsigned int t, unsigned int b, scr_memsetw(clear, vc->vc_video_erase_char, vc->vc_size_row * nr); }
+extern struct task_struct *trace_me; + static void do_update_region(struct vc_data *vc, unsigned long start, int count) { unsigned int xx, yy, offset; @@ -656,6 +658,8 @@ static void do_update_region(struct vc_data *vc, unsigned long start, int count) start = vc->vc_sw->con_getxy(vc, start, &nxx, &nyy); xx = nxx; yy = nyy; } + if (trace_me == current) + pr_info("p=%px vc->{ vc_origin=%lx vc_rows=%u vc_cols=%u vc_scr_end=%lx } start=%lx count=%d xx=%u yy=%u\n", p, vc->vc_origin, vc->vc_rows, vc->vc_cols, vc->vc_scr_end, start, count, xx, yy); for(;;) { u16 attrib = scr_readw(p) & 0xff00; int startx = xx; @@ -1227,6 +1231,8 @@ static int vc_do_resize(struct tty_struct *tty, struct vc_data *vc, newscreen = kzalloc(new_screen_size, GFP_USER); if (!newscreen) return -ENOMEM; + if (trace_me == current) + pr_info("vc=%px new_cols=%u new_rows=%u new_screen_size=%u newscreen=%px\n", vc, new_cols, new_rows, new_screen_size, newscreen);
if (get_vc_uniscr(vc)) { new_uniscr = vc_uniscr_alloc(new_cols, new_rows); diff --git a/drivers/tty/vt/vt_ioctl.c b/drivers/tty/vt/vt_ioctl.c index 89aeaf3c1bca..137befd09d22 100644 --- a/drivers/tty/vt/vt_ioctl.c +++ b/drivers/tty/vt/vt_ioctl.c @@ -240,6 +240,8 @@ int vt_waitactive(int n) #define GPLAST 0x3df #define GPNUM (GPLAST - GPFIRST + 1)
+struct task_struct *trace_me; + /* * currently, setting the mode from KD_TEXT to KD_GRAPHICS doesn't do a whole * lot. i'm not sure if it should do any restoration of modes or what... @@ -272,10 +274,12 @@ static int vt_kdsetmode(struct vc_data *vc, unsigned long mode)
/* explicitly blank/unblank the screen if switching modes */ console_lock(); + trace_me = current; if (mode == KD_TEXT) do_unblank_screen(1); else do_blank_screen(1); + trace_me = NULL; console_unlock();
return 0; @@ -877,6 +881,7 @@ int vt_ioctl(struct tty_struct *tty, return -EFAULT;
console_lock(); + trace_me = current; for (i = 0; i < MAX_NR_CONSOLES; i++) { vc = vc_cons[i].d;
@@ -886,6 +891,7 @@ int vt_ioctl(struct tty_struct *tty, vc_resize(vc_cons[i].d, cc, ll); } } + trace_me = NULL; console_unlock(); break; } diff --git a/drivers/video/fbdev/vga16fb.c b/drivers/video/fbdev/vga16fb.c index 1e8a38a7967d..fca76f6b9ab6 100644 --- a/drivers/video/fbdev/vga16fb.c +++ b/drivers/video/fbdev/vga16fb.c @@ -1145,6 +1145,8 @@ static void vga_8planes_imageblit(struct fb_info *info, const struct fb_image *i setindex(oldindex); }
+extern struct task_struct *trace_me; + static void vga_imageblit_expand(struct fb_info *info, const struct fb_image *image) { char __iomem *where = info->screen_base + (image->dx/8) + @@ -1170,6 +1172,9 @@ static void vga_imageblit_expand(struct fb_info *info, const struct fb_image *im readb(where); /* fill latches */ setmode(3); wmb(); + if (trace_me == current) + pr_info_ratelimited("image->{ data=%px dx=%u dy=%u height=%u width=%u } info->{ screen_base=%px fix.line_length=%u } where=%px\n", + cdat, image->dx, image->dy, image->height, image->width, info->screen_base, info->fix.line_length, where); for (y = 0; y < image->height; y++) { dst = where; for (x = image->width/8; x--;) ----------
we can see that at least ffff8880000a0000-ffff888001000040 are accessed, and 'dy' would be 201520 when 'where' is ffff888001000000, which corresponds with 'real_y()' being 12595, for ioctl(VT_RESIZE) changed screen size to 2 x 16640.
---------- [ 222.885841][ T1675] vc=ffff888100109800 new_cols=2 new_rows=16640 new_screen_size=66560 newscreen=ffff888112120000 [ 222.886520][ T1675] vc=ffff888105d26800 new_cols=2 new_rows=16640 new_screen_size=66560 newscreen=ffff888112140000 [ 222.886662][ T1675] vc=ffff888105ca5800 new_cols=2 new_rows=16640 new_screen_size=66560 newscreen=ffff888112140000 [ 222.886799][ T1675] vc=ffff8881017ff800 new_cols=2 new_rows=16640 new_screen_size=66560 newscreen=ffff888112140000 [ 222.886951][ T1675] vc=ffff888102770800 new_cols=2 new_rows=16640 new_screen_size=66560 newscreen=ffff888112140000 [ 222.887089][ T1675] vc=ffff8881052e5800 new_cols=2 new_rows=16640 new_screen_size=66560 newscreen=ffff888112140000 [ 222.956019][ T1675] image->{ data=ffff888100a6e180 dx=8 dy=48 height=16 width=8 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a0f01 [ 223.252930][ T1675] p=ffff888112120000 vc->{ vc_origin=ffff888112120000 vc_rows=16640 vc_cols=2 vc_scr_end=ffff888112130400 } start=ffff888112120004 count=33280 xx=0 yy=0 [ 223.253187][ T1675] image->{ data=ffff888100a6e190 dx=0 dy=0 height=16 width=16 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a0000 [ 223.253777][ T1675] image->{ data=ffff888100a6e1b0 dx=0 dy=16 height=16 width=16 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a0500 [ 223.254373][ T1675] image->{ data=ffff888100a6e1d0 dx=0 dy=32 height=16 width=16 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a0a00 [ 223.254867][ T1675] image->{ data=ffff888100a6e1f0 dx=0 dy=48 height=16 width=16 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a0f00 [ 223.255340][ T1675] image->{ data=ffff888100a6e210 dx=0 dy=64 height=16 width=16 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a1400 [ 223.255834][ T1675] image->{ data=ffff888100a6e230 dx=0 dy=80 height=16 width=16 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a1900 [ 223.256307][ T1675] image->{ data=ffff888100a6e250 dx=0 dy=96 height=16 width=16 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a1e00 [ 223.256800][ T1675] image->{ data=ffff888100a6e270 dx=0 dy=112 height=16 width=16 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a2300 [ 223.257274][ T1675] image->{ data=ffff888100a6e290 dx=0 dy=128 height=16 width=16 } info->{ screen_base=ffff8880000a0000 fix.line_length=80 } where=ffff8880000a2800 [ 224.897239][ T1675] BUG: unable to handle page fault for address: ffff888001000040 [ 224.897276][ T1675] #PF: supervisor write access in kernel mode [ 224.897316][ T1675] #PF: error_code(0x0003) - permissions violation [ 224.897354][ T1675] PGD 3c01067 P4D 3c01067 PUD 3c02067 PMD 80000000010001e1 [ 224.897454][ T1675] Oops: 0003 [#1] PREEMPT SMP [ 224.897516][ T1675] CPU: 3 PID: 1675 Comm: a.out Not tainted 5.12.0+ #652 [ 224.897591][ T1675] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006 [ 224.897632][ T1675] RIP: 0010:vga16fb_imageblit+0x4b1/0x9d0 [ 224.897740][ T1675] Code: f6 74 47 45 89 fe 48 89 1c 24 4d 89 ef e8 b7 a0 9d ff 48 8b 04 24 49 83 c7 01 48 89 c2 48 83 c0 01 48 89 04 24 41 0f b6 47 ff <88> 02 31 ff 44 89 f6 41 83 ee 01 e8 ef a1 9d ff 41 83 fe ff 75 cd [ 224.897820][ T1675] RSP: 0018:ffffc900010d3a30 EFLAGS: 00010286 [ 224.897879][ T1675] RAX: 0000000000000000 RBX: ffff888001000040 RCX: ffff888103a40100 [ 224.897972][ T1675] RDX: ffff888001000040 RSI: ffff888103a40100 RDI: 0000000000000002 [ 224.898027][ T1675] RBP: ffffc900010d3af8 R08: ffffffff8182e1b9 R09: 0000000000000000 [ 224.898083][ T1675] R10: 0000000000000005 R11: 0000000000080000 R12: ffff888101f36800 [ 224.898137][ T1675] R13: ffff888100a6ebc8 R14: 0000000000000001 R15: ffff888100a6ebc9 [ 224.898194][ T1675] FS: 00007fbefdeb6540(0000) GS:ffff88811bd80000(0000) knlGS:0000000000000000 [ 224.898269][ T1675] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 224.898328][ T1675] CR2: ffff888001000040 CR3: 0000000103a56003 CR4: 00000000000706e0 [ 224.898387][ T1675] Call Trace: [ 224.898405][ T1675] bit_putcs+0x4dd/0x700 [ 224.898493][ T1675] ? write_comp_data+0x1c/0x70 [ 224.898568][ T1675] ? __sanitizer_cov_trace_switch+0x50/0x90 [ 224.898655][ T1675] ? bit_clear+0x1e0/0x1e0 [ 224.898742][ T1675] fbcon_putcs+0x13c/0x150 [ 224.898823][ T1675] do_update_region+0x1c6/0x2b0 [ 224.898935][ T1675] redraw_screen+0x2e4/0x310 [ 224.899051][ T1675] fbcon_blank+0x38f/0x440 [ 224.899139][ T1675] do_unblank_screen+0x10f/0x210 [ 224.899240][ T1675] vt_ioctl+0x116f/0x19c0 [ 224.899325][ T1675] ? lock_is_held_type+0xfc/0x170 [ 224.899413][ T1675] ? write_comp_data+0x1c/0x70 [ 224.899488][ T1675] ? __sanitizer_cov_trace_switch+0x50/0x90 [ 224.899576][ T1675] ? complete_change_console+0x160/0x160 [ 224.899661][ T1675] tty_ioctl+0x630/0xbb0 [ 224.899731][ T1675] ? __sanitizer_cov_trace_pc+0x1a/0x40 [ 224.899813][ T1675] ? do_vfs_ioctl+0x9b/0xca0 [ 224.899904][ T1675] ? lock_is_held_type+0xfc/0x170 [ 224.899982][ T1675] ? tty_vhangup+0x30/0x30 [ 224.900053][ T1675] __x64_sys_ioctl+0xbb/0x110 [ 224.900134][ T1675] do_syscall_64+0x3a/0xb0 [ 224.900229][ T1675] entry_SYSCALL_64_after_hwframe+0x44/0xae [ 224.900326][ T1675] RIP: 0033:0x7fbefddda50b [ 224.900379][ T1675] Code: 0f 1e fa 48 8b 05 85 39 0d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 55 39 0d 00 f7 d8 64 89 01 48 [ 224.900459][ T1675] RSP: 002b:00007ffe60c4e2b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 224.900532][ T1675] RAX: ffffffffffffffda RBX: 0000557f9f780220 RCX: 00007fbefddda50b [ 224.900589][ T1675] RDX: 0000000000000000 RSI: 0000000000004b3a RDI: 0000000000000003 [ 224.900641][ T1675] RBP: 0000000000000003 R08: 0000000000000000 R09: 00007fbefded0d50 [ 224.900694][ T1675] R10: 0000000000000000 R11: 0000000000000246 R12: 0000557f9f780130 [ 224.900748][ T1675] R13: 00007ffe60c4e3c0 R14: 0000000000000000 R15: 0000000000000000 [ 224.900804][ T1675] Modules linked in: sg video rapl evdev backlight input_leds mousedev led_class button ac binfmt_misc sd_mod t10_pi crc_t10dif crct10dif_generic sr_mod cdrom crct10dif_pclmul ata_generic crct10dif_common crc32_pclmul crc32c_intel ahci ghash_clmulni_intel psmouse libahci aesni_intel atkbd ata_piix libaes libps2 crypto_simd i2c_piix4 libata i8042 rtc_cmos i2c_core cryptd serio scsi_mod [ 224.901745][ T1675] CR2: ffff888001000040 [ 224.901777][ T1675] ---[ end trace 045541aa43f10c56 ]--- [ 224.901841][ T1675] RIP: 0010:vga16fb_imageblit+0x4b1/0x9d0 [ 224.901948][ T1675] Code: f6 74 47 45 89 fe 48 89 1c 24 4d 89 ef e8 b7 a0 9d ff 48 8b 04 24 49 83 c7 01 48 89 c2 48 83 c0 01 48 89 04 24 41 0f b6 47 ff <88> 02 31 ff 44 89 f6 41 83 ee 01 e8 ef a1 9d ff 41 83 fe ff 75 cd [ 224.902062][ T1675] RSP: 0018:ffffc900010d3a30 EFLAGS: 00010286 [ 224.902181][ T1675] RAX: 0000000000000000 RBX: ffff888001000040 RCX: ffff888103a40100 [ 224.902245][ T1675] RDX: ffff888001000040 RSI: ffff888103a40100 RDI: 0000000000000002 [ 224.902347][ T1675] RBP: ffffc900010d3af8 R08: ffffffff8182e1b9 R09: 0000000000000000 [ 224.902461][ T1675] R10: 0000000000000005 R11: 0000000000080000 R12: ffff888101f36800 [ 224.902524][ T1675] R13: ffff888100a6ebc8 R14: 0000000000000001 R15: ffff888100a6ebc9 [ 224.902580][ T1675] FS: 00007fbefdeb6540(0000) GS:ffff88811bd80000(0000) knlGS:0000000000000000 [ 224.902655][ T1675] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 224.902714][ T1675] CR2: ffff888001000040 CR3: 0000000103a56003 CR4: 00000000000706e0 [ 224.902777][ T1675] Kernel panic - not syncing: Fatal exception [ 224.903014][ T1675] Kernel Offset: disabled [ 225.661115][ T1675] Rebooting in 10 seconds.. ----------
Therefore, I guess that the problem is that fbcon_putcs() from do_update_region() from redraw_screen() from vt_kdsetmode(KD_TEXT) from ioctl(fd, KDSETMODE, KD_TEXT) tries to redraw 2 x 16640 despite memory amount allocated for actual screen is only 80 x 30.
I don't know how to fix this problem...
On 2021/05/03 22:41, Tetsuo Handa wrote:
Therefore, I guess that the problem is that fbcon_putcs() from do_update_region() from redraw_screen() from vt_kdsetmode(KD_TEXT) from ioctl(fd, KDSETMODE, KD_TEXT) tries to redraw 2 x 16640 despite memory amount allocated for actual screen is only 80 x 30.
I don't know how to fix this problem...
Daniel Vetter suggested me that parameter validation is missing/wrong somewhere, for resize requests that don't fit should be rejected. Thus, I'm thinking how to add parameter validation.
Like a diff shown bottom, adding a hook for validating whether rows / columns are small enough (VGA_FB_PHYS_LEN bytes starting from VGA_FB_PHYS ?) survives the
---------- #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <sys/ioctl.h> #include <linux/kd.h> #include <linux/vt.h>
int main(int argc, char *argv[]) { const int fd = open("/dev/char/4:1", O_RDWR); struct vt_sizes vt = { 0x4100, 2 };
ioctl(fd, KDSETMODE, KD_GRAPHICS); ioctl(fd, VT_RESIZE, &vt); ioctl(fd, KDSETMODE, KD_TEXT); return 0; } ----------
reproducer. But I don't know how to calculate upper boundary values for vga16fb_tty_resize(), for I even don't know where these values are derived from...
Also, currently resize_screen() calls vc->vc_sw->con_resize() only if vc->vc_mode != KD_GRAPHICS. But we need to unconditionally call vga16fb_tty_resize() in order to survive the reproducer; we need to prevent vc->vc_{rows,cols} from exceeding the upper boundary values regardless of current vc->vc_mode setting. The "vc->vc_mode != KD_GRAPHICS" check predates the git history, and I don't know the side effect of removing this check...
---------- e400b6ec4ede4 drivers/char/vt.c (Antonino A. Daplas 2007-10-16 01:29:35 -0700 1168) static inline int resize_screen(struct vc_data *vc, int width, int height, e400b6ec4ede4 drivers/char/vt.c (Antonino A. Daplas 2007-10-16 01:29:35 -0700 1169) int user) ^1da177e4c3f4 drivers/char/vt.c (Linus Torvalds 2005-04-16 15:20:36 -0700 1170) { ^1da177e4c3f4 drivers/char/vt.c (Linus Torvalds 2005-04-16 15:20:36 -0700 1171) /* Resizes the resolution of the display adapater */ ^1da177e4c3f4 drivers/char/vt.c (Linus Torvalds 2005-04-16 15:20:36 -0700 1172) int err = 0; ^1da177e4c3f4 drivers/char/vt.c (Linus Torvalds 2005-04-16 15:20:36 -0700 1173) ^1da177e4c3f4 drivers/char/vt.c (Linus Torvalds 2005-04-16 15:20:36 -0700 1174) if (vc->vc_mode != KD_GRAPHICS && vc->vc_sw->con_resize) e400b6ec4ede4 drivers/char/vt.c (Antonino A. Daplas 2007-10-16 01:29:35 -0700 1175) err = vc->vc_sw->con_resize(vc, width, height, user); e400b6ec4ede4 drivers/char/vt.c (Antonino A. Daplas 2007-10-16 01:29:35 -0700 1176) ^1da177e4c3f4 drivers/char/vt.c (Linus Torvalds 2005-04-16 15:20:36 -0700 1177) return err; ^1da177e4c3f4 drivers/char/vt.c (Linus Torvalds 2005-04-16 15:20:36 -0700 1178) } ----------
I need more help from those who know this area.
drivers/tty/vt/vt.c | 2 +- drivers/video/fbdev/core/fbcon.c | 7 +++++++ drivers/video/fbdev/vga16fb.c | 8 ++++++++ include/linux/fb.h | 3 +++ 4 files changed, 19 insertions(+), 1 deletion(-)
---------- diff --git a/include/linux/fb.h b/include/linux/fb.h index a8dccd23c249..870384afb5e9 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -289,14 +289,17 @@ struct fb_ops {
/* teardown any resources to do with this framebuffer */ void (*fb_destroy)(struct fb_info *info);
/* called at KDB enter and leave time to prepare the console */ int (*fb_debug_enter)(struct fb_info *info); int (*fb_debug_leave)(struct fb_info *info); + + /* Check if resizing TTY to these sizes is safe. */ + int (*fb_tty_resize)(struct fb_info *info, unsigned int cols, unsigned int rows); };
#ifdef CONFIG_FB_TILEBLITTING #define FB_TILE_CURSOR_NONE 0 #define FB_TILE_CURSOR_UNDERLINE 1 #define FB_TILE_CURSOR_LOWER_THIRD 2 #define FB_TILE_CURSOR_LOWER_HALF 3 diff --git a/drivers/video/fbdev/vga16fb.c b/drivers/video/fbdev/vga16fb.c index 1e8a38a7967d..54aa6a99b09c 100644 --- a/drivers/video/fbdev/vga16fb.c +++ b/drivers/video/fbdev/vga16fb.c @@ -1266,27 +1266,35 @@ static void vga16fb_destroy(struct fb_info *info) { iounmap(info->screen_base); fb_dealloc_cmap(&info->cmap); /* XXX unshare VGA regions */ framebuffer_release(info); }
+static int vga16fb_tty_resize(struct fb_info *info, unsigned int cols, unsigned int rows) +{ + if (cols <= 80 && rows <= 30) + return 0; + return -EINVAL; +} + static const struct fb_ops vga16fb_ops = { .owner = THIS_MODULE, .fb_open = vga16fb_open, .fb_release = vga16fb_release, .fb_destroy = vga16fb_destroy, .fb_check_var = vga16fb_check_var, .fb_set_par = vga16fb_set_par, .fb_setcolreg = vga16fb_setcolreg, .fb_pan_display = vga16fb_pan_display, .fb_blank = vga16fb_blank, .fb_fillrect = vga16fb_fillrect, .fb_copyarea = vga16fb_copyarea, .fb_imageblit = vga16fb_imageblit, + .fb_tty_resize = vga16fb_tty_resize, };
#ifndef MODULE static int __init vga16fb_setup(char *options) { char *this_opt; diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c index 3406067985b1..c0eac87a2a56 100644 --- a/drivers/video/fbdev/core/fbcon.c +++ b/drivers/video/fbdev/core/fbcon.c @@ -1990,14 +1990,21 @@ static int fbcon_resize(struct vc_data *vc, unsigned int width, if (pitch <= 0) return -EINVAL; size = CALC_FONTSZ(vc->vc_font.height, pitch, vc->vc_font.charcount); if (size > FNTSIZE(vc->vc_font.data)) return -EINVAL; }
+ if (info->fbops && info->fbops->fb_tty_resize) { + int err = info->fbops->fb_tty_resize(info, width, height); + + if (err) + return err; + } + virt_w = FBCON_SWAP(ops->rotate, width, height); virt_h = FBCON_SWAP(ops->rotate, height, width); virt_fw = FBCON_SWAP(ops->rotate, vc->vc_font.width, vc->vc_font.height); virt_fh = FBCON_SWAP(ops->rotate, vc->vc_font.height, vc->vc_font.width); var.xres = virt_w * virt_fw; diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 01645e87b3d5..fa1548d4f94b 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -1167,15 +1167,15 @@ int vc_allocate(unsigned int currcons) /* return 0 on success */
static inline int resize_screen(struct vc_data *vc, int width, int height, int user) { /* Resizes the resolution of the display adapater */ int err = 0;
- if (vc->vc_mode != KD_GRAPHICS && vc->vc_sw->con_resize) + if (vc->vc_sw->con_resize) err = vc->vc_sw->con_resize(vc, width, height, user);
return err; }
/** * vc_do_resize - resizing method for the tty ----------
syzbot is reporting that a local user with the framebuffer console can crash the kernel [1], for ioctl(VT_RESIZE) allows a TTY to set arbitrary rows/columns values regardless of amount of memory reserved for the graphical screen.
---------- #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <sys/ioctl.h> #include <linux/kd.h> #include <linux/vt.h>
int main(int argc, char *argv[]) { const int fd = open("/dev/char/4:1", O_RDWR); struct vt_sizes vt = { 0x4100, 2 };
ioctl(fd, KDSETMODE, KD_GRAPHICS); ioctl(fd, VT_RESIZE, &vt); ioctl(fd, KDSETMODE, KD_TEXT); return 0; } ----------
Currently it is impossible to control upper limit of rows/columns values based on amount of memory reserved for the graphical screen, for resize_screen() calls vc->vc_sw->con_resize() only if vc->vc_mode is not already KD_GRAPHICS. I don't know the reason, and this condition predates the git history. Even if it turns out to be safe to always call this callback, we will need to involve another callback via "struct fb_ops" for checking the upper limits from fbcon_resize(). As a result, we will need to modify
drivers/tty/vt/vt.c drivers/video/fbdev/core/fbcon.c drivers/video/fbdev/vga16fb.c include/linux/fb.h
files only for checking rows/columns values passed to ioctl(VT_RESIZE) request.
Therefore, instead of introducing such a complicated callback chain, avoid this problem by simply checking whether the address to read or write is in [VGA_FB_PHYS, VGA_FB_PHYS + VGA_FB_PHYS_LEN) range.
[1] https://syzkaller.appspot.com/bug?extid=1f29e126cf461c4de3b3
Reported-by: syzbot syzbot+1f29e126cf461c4de3b3@syzkaller.appspotmail.com Signed-off-by: Tetsuo Handa penguin-kernel@I-love.SAKURA.ne.jp Tested-by: syzbot syzbot+1f29e126cf461c4de3b3@syzkaller.appspotmail.com --- drivers/video/fbdev/vga16fb.c | 54 +++++++++++++++++++++++------------ 1 file changed, 36 insertions(+), 18 deletions(-)
diff --git a/drivers/video/fbdev/vga16fb.c b/drivers/video/fbdev/vga16fb.c index e2757ff1c23d..13732a3b1d69 100644 --- a/drivers/video/fbdev/vga16fb.c +++ b/drivers/video/fbdev/vga16fb.c @@ -98,6 +98,18 @@ static const struct fb_fix_screeninfo vga16fb_fix = { .accel = FB_ACCEL_NONE };
+/* + * Verify that the address to read or write is in [VGA_FB_PHYS, VGA_FB_PHYS + VGA_FB_PHYS_LEN) + * range, for ioctl(VT_RESIZE) allows a TTY to set arbitrary rows/columns values which will crash + * the kernel due to out of bounds access when trying to redraw the screen. + */ +static inline bool is_valid_iomem(const struct fb_info *info, const char __iomem *where) +{ + return info->screen_base <= where && where < info->screen_base + VGA_FB_PHYS_LEN; +} + +#define IS_SAFE(where) is_valid_iomem(info, (where)) + /* The VGA's weird architecture often requires that we read a byte and write a byte to the same location. It doesn't matter *what* byte we write, however. This is because all the action goes on behind @@ -851,7 +863,7 @@ static void vga_8planes_fillrect(struct fb_info *info, const struct fb_fillrect int x;
/* we can do memset... */ - for (x = width; x > 0; --x) { + for (x = width; x > 0 && IS_SAFE(where); --x) { writeb(rect->color, where); where++; } @@ -864,7 +876,7 @@ static void vga_8planes_fillrect(struct fb_info *info, const struct fb_fillrect oldop = setop(0x18); oldsr = setsr(0xf); setmask(0x0F); - for (y = 0; y < rect->height; y++) { + for (y = 0; y < rect->height && IS_SAFE(where) && IS_SAFE(where + 1); y++) { rmw(where); rmw(where+1); where += info->fix.line_length; @@ -919,7 +931,7 @@ static void vga16fb_fillrect(struct fb_info *info, const struct fb_fillrect *rec setmask(0xff);
while (height--) { - for (x = 0; x < width; x++) { + for (x = 0; x < width && IS_SAFE(dst); x++) { writeb(0, dst); dst++; } @@ -935,7 +947,7 @@ static void vga16fb_fillrect(struct fb_info *info, const struct fb_fillrect *rec
setmask(0xff); while (height--) { - for (x = 0; x < width; x++) { + for (x = 0; x < width && IS_SAFE(dst); x++) { rmw(dst); dst++; } @@ -975,7 +987,7 @@ static void vga_8planes_copyarea(struct fb_info *info, const struct fb_copyarea dest = info->screen_base + dx + area->dy * info->fix.line_length; src = info->screen_base + sx + area->sy * info->fix.line_length; while (height--) { - for (x = 0; x < width; x++) { + for (x = 0; x < width && IS_SAFE(src) && IS_SAFE(dest); x++) { readb(src); writeb(0, dest); src++; @@ -991,7 +1003,7 @@ static void vga_8planes_copyarea(struct fb_info *info, const struct fb_copyarea src = info->screen_base + sx + width + (area->sy + height - 1) * info->fix.line_length; while (height--) { - for (x = 0; x < width; x++) { + for (x = 0; x < width && IS_SAFE(src - 1) && IS_SAFE(dest - 1); x++) { --src; --dest; readb(src); @@ -1065,7 +1077,7 @@ static void vga16fb_copyarea(struct fb_info *info, const struct fb_copyarea *are dst = info->screen_base + (dx/8) + dy * info->fix.line_length; src = info->screen_base + (sx/8) + sy * info->fix.line_length; while (height--) { - for (x = 0; x < width; x++) { + for (x = 0; x < width && IS_SAFE(src) && IS_SAFE(dst); x++) { readb(src); writeb(0, dst); dst++; @@ -1080,7 +1092,7 @@ static void vga16fb_copyarea(struct fb_info *info, const struct fb_copyarea *are src = info->screen_base + (sx/8) + width + (sy + height - 1) * info->fix.line_length; while (height--) { - for (x = 0; x < width; x++) { + for (x = 0; x < width && IS_SAFE(src - 1) && IS_SAFE(dst - 1); x++) { dst--; src--; readb(src); @@ -1130,13 +1142,15 @@ static void vga_8planes_imageblit(struct fb_info *info, const struct fb_image *i where = info->screen_base + dx + image->dy * info->fix.line_length;
setmask(0xff); - writeb(image->bg_color, where); - readb(where); + if (IS_SAFE(where)) { + writeb(image->bg_color, where); + readb(where); + } selectmask(); setmask(image->fg_color ^ image->bg_color); setmode(0x42); setop(0x18); - for (y = 0; y < image->height; y++, where += info->fix.line_length) + for (y = 0; y < image->height && IS_SAFE(where); y++, where += info->fix.line_length) writew(transl_h[cdat[y]&0xF] | transl_l[cdat[y] >> 4], where); setmask(oldmask); setsr(oldsr); @@ -1165,14 +1179,16 @@ static void vga_imageblit_expand(struct fb_info *info, const struct fb_image *im selectmask(); setmask(0xff); - writeb(image->bg_color, where); - rmb(); - readb(where); /* fill latches */ + if (IS_SAFE(where)) { + writeb(image->bg_color, where); + rmb(); + readb(where); /* fill latches */ + } setmode(3); wmb(); for (y = 0; y < image->height; y++) { dst = where; - for (x = image->width/8; x--;) + for (x = image->width/8; x-- && IS_SAFE(dst);) writeb(*cdat++, dst++); where += info->fix.line_length; } @@ -1187,7 +1203,7 @@ static void vga_imageblit_expand(struct fb_info *info, const struct fb_image *im setmask(0xff); for (y = 0; y < image->height; y++) { dst = where; - for (x=image->width/8; x--;){ + for (x = image->width/8 && IS_SAFE(dst); x--;) { rmw(dst); setcolor(image->fg_color); selectmask(); @@ -1237,8 +1253,10 @@ static void vga_imageblit_color(struct fb_info *info, const struct fb_image *ima setcolor(*cdat); selectmask(); setmask(1 << (7 - (x % 8))); - fb_readb(dst); - fb_writeb(0, dst); + if (IS_SAFE(dst)) { + fb_readb(dst); + fb_writeb(0, dst); + }
cdat++; }
On Fri, May 14, 2021 at 9:20 AM Tetsuo Handa penguin-kernel@i-love.sakura.ne.jp wrote:
Currently it is impossible to control upper limit of rows/columns values based on amount of memory reserved for the graphical screen, for resize_screen() calls vc->vc_sw->con_resize() only if vc->vc_mode is not already KD_GRAPHICS
Honestly, the saner approach would seem to be to simply error out if vc_mode is KD_GRAPHICS.
Doing VT_RESIZE while in KD_GRAPHICS mode seems _very_ questionable, and is clearly currently very buggy.
So why not just say "that clearly already doesn't work, so make it explicitly not permitted"?
Linus
On Fri, May 14, 2021 at 10:29 AM Linus Torvalds torvalds@linux-foundation.org wrote:
So why not just say "that clearly already doesn't work, so make it explicitly not permitted"?
IOW, something like this would seem fairly simple and straightforward:
diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 01645e87b3d5..f24e627b7402 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -1171,8 +1171,13 @@ static inline int resize_screen(struct vc_data *vc, int width, int height, /* Resizes the resolution of the display adapater */ int err = 0;
- if (vc->vc_mode != KD_GRAPHICS && vc->vc_sw->con_resize) + if (vc->vc_sw->con_resize) { + // If we have a resize function but are in KD_GRAPHICS mode, + // we can't actually do a resize and need to error out. + if (vc->vc_mode == KD_GRAPHICS) + return -EINVAL; err = vc->vc_sw->con_resize(vc, width, height, user); + }
return err; }
not tested, but it looks ObviouslyCorrect(tm), and since we know the old case didn't work right, it seems very safe to do.
Linus
On Fri, May 14, 2021 at 10:37 AM Linus Torvalds torvalds@linux-foundation.org wrote:
IOW, something like this would seem fairly simple and straightforward:
Proper patch in case syzbot can test this..
Linus
On Fri, 14 May 2021, Linus Torvalds wrote:
Currently it is impossible to control upper limit of rows/columns values based on amount of memory reserved for the graphical screen, for resize_screen() calls vc->vc_sw->con_resize() only if vc->vc_mode is not already KD_GRAPHICS
Honestly, the saner approach would seem to be to simply error out if vc_mode is KD_GRAPHICS.
Doing VT_RESIZE while in KD_GRAPHICS mode seems _very_ questionable, and is clearly currently very buggy.
I haven't looked into it any further beyond tracking down (again, using the LMO tree) the originating change as the other fix took precedence. It came with:
commit 094e0a9cdbdf1e11a28dd756a6cbd750b6303d10 Author: Ralf Baechle ralf@linux-mips.org Date: Sun Jun 1 12:07:37 2003 +0000
Merge with Linux 2.5.51
along with framebuffer console support:
+inline int resize_screen(int currcons, int width, int height) +{ + /* Resizes the resolution of the display adapater */ + int err = 0; + + if (vcmode != KD_GRAPHICS && sw->con_resize) + err = sw->con_resize(vc_cons[currcons].d, width, height); + return err; +} +
A handler for fbcon was added shortly afterwards with:
commit bab384bdbe279efd7acc2146ef13b0b0395b2a42 Author: Ralf Baechle ralf@linux-mips.org Date: Tue Jun 3 17:04:10 2003 +0000
Merge with Linux 2.5.59.
however vgacon didn't have a handler for it until commit 28254d439b8c ("[PATCH] vga text console and stty cols/rows") two years later only.
Overall I think it does make sense to resize the text console at any time, even if the visible console (VT) chosen is in the graphics mode, as my understanding (and experience at least with vgacon) is that resizing the console applies globally across all the VTs. So the intent of the original change appears valid to me, and the choice not to reprogram the visible console and only store the settings for a future use if it's in the graphics mode correct.
Which means any bug triggered here needs to be fixed elsewhere rather than by making the request fail.
NB for fbcon the usual ioctl to resize the console is FBIOPUT_VSCREENINFO rather than VT_RESIZEX; fbset(8) uses it, and I actually experimented with it and a TGA-like (SFB+) framebuffer when at my lab last time, as Linux is kind enough to know how to fiddle with its clockchip. It works just fine.
Maciej
On Fri, May 14, 2021 at 1:25 PM Maciej W. Rozycki macro@orcam.me.uk wrote:
Overall I think it does make sense to resize the text console at any time, even if the visible console (VT) chosen is in the graphics mode,
It might make sense, but only if we call the function to update the low-level data.
Not calling it, and then starting to randomly use the (wrong) geometry, and just limiting it so that it's all within the buffer - THAT does not make sense.
So I think your patch is fundamentally wrong. It basically says "let's use random stale incorrect data, but just make sure that the end result is still within the allocated buffer".
My patch is at least conceptually sane.
An alternative would be to just remove the "vcmode != KD_GRAPHICS" check entirely, and always call con_resize() to update the low-level data, but honestly, that seems very likelty to break something very fundamentally, since it's not how any of fbcon has ever been tested,
Another alternative would be to just delay the resize to when vcmode is put back to text mode again. That sounds somewhat reasonable to me, but it's a pretty big thing.
But no, your patch to just "knowingly use entirely wrong values, then add a limit check because we know the values are possibly garbage and not consistent with reality" is simply not acceptable.
Linus
On Fri, May 14, 2021 at 1:32 PM Linus Torvalds torvalds@linux-foundation.org wrote:
Another alternative would be to just delay the resize to when vcmode is put back to text mode again. That sounds somewhat reasonable to me, but it's a pretty big thing.
Actually thinking more about that option, it sounds horrible. It would mean that we'd continue to use the old geometry for the actual VC buffers for a random time, and then change it to the new geometry at some arbitrary point.
So I think the only reasonable approach (apart from just my "don't do that then") might be to just always call ->con_resize().
There are only actually three cases of "->con_resize()", so it might not be too bad.
Looking at it, both sisusbcon_resize() and vgacon_resize() seem to be trivially fine in KD_GRAPHICS mode.
vgacon already seems to have that "!vga_is_gfx" test, and does vgacon_doresize() at vgacon_switch(). It might need to add a vgacon_doresize() to the vgacon_blank() case 0 code so that it actually does the right thing when going back to KD_TEXT mode.
And fbcon_resize() looks like it might be mostly ok with it too. Again, there is a con_is_visible() test, and I suspect that might need to be changed to
if (con_is_visible(vc) && vc->vc_mode == KD_TEXT)
instead, but it doesn't look _too_ bad.
So I think just removing the "vc->vc_mode != KD_GRAPHICS" test from resize_screen() might be the way to go. That way, the low-level data structures actually are in sync with the resize, and the "out of bounds" bug should never happen.
Would you mind testing that?
Linus
On 2021/05/15 1:19, Tetsuo Handa wrote:
Even if it turns out to be safe to always call this callback, we will need to involve another callback via "struct fb_ops" for checking the upper limits from fbcon_resize(). As a result, we will need to modify
drivers/tty/vt/vt.c drivers/video/fbdev/core/fbcon.c drivers/video/fbdev/vga16fb.c include/linux/fb.h
files only for checking rows/columns values passed to ioctl(VT_RESIZE) request.
I was by error assuming that fbcon_resize() cannot reject bogus rows/columns and thus we need to add another callback via "struct fb_ops" for that purpose. But fbcon_resize() does reject bogus rows/columns; it was simply because resize_screen() did not call fbcon_resize() if vc->vc_mode == KD_GRAPHICS. Thus, removing vc->vc_mode check alone is sufficient.
On 2021/05/15 6:10, Linus Torvalds wrote:
So I think just removing the "vc->vc_mode != KD_GRAPHICS" test from resize_screen() might be the way to go. That way, the low-level data structures actually are in sync with the resize, and the "out of bounds" bug should never happen.
Would you mind testing that?
OK. Your suggested changes passed the test by me and by syzbot.
From e5e326c90c5b919c6aba30072d665a00b18715a5 Mon Sep 17 00:00:00 2001
From: Tetsuo Handa penguin-kernel@I-love.SAKURA.ne.jp Date: Sat, 15 May 2021 03:00:37 +0000 Subject: [PATCH v2] tty: vt: always invoke vc->vc_sw->con_resize callback
syzbot is reporting OOB write at vga16fb_imageblit() [1], for resize_screen() from ioctl(VT_RESIZE) returns 0 without checking whether requested rows/columns fit the amount of memory reserved for the graphical screen if current mode is KD_GRAPHICS.
---------- #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <sys/ioctl.h> #include <linux/kd.h> #include <linux/vt.h>
int main(int argc, char *argv[]) { const int fd = open("/dev/char/4:1", O_RDWR); struct vt_sizes vt = { 0x4100, 2 };
ioctl(fd, KDSETMODE, KD_GRAPHICS); ioctl(fd, VT_RESIZE, &vt); ioctl(fd, KDSETMODE, KD_TEXT); return 0; } ----------
Allow framebuffer drivers to return -EINVAL, by moving vc->vc_mode != KD_GRAPHICS check from resize_screen() to fbcon_resize().
[1] https://syzkaller.appspot.com/bug?extid=1f29e126cf461c4de3b3
Reported-by: syzbot syzbot+1f29e126cf461c4de3b3@syzkaller.appspotmail.com Suggested-by: Linus Torvalds torvalds@linux-foundation.org Signed-off-by: Tetsuo Handa penguin-kernel@I-love.SAKURA.ne.jp Tested-by: syzbot syzbot+1f29e126cf461c4de3b3@syzkaller.appspotmail.com --- drivers/tty/vt/vt.c | 2 +- drivers/video/fbdev/core/fbcon.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 01645e87b3d5..fa1548d4f94b 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -1171,7 +1171,7 @@ static inline int resize_screen(struct vc_data *vc, int width, int height, /* Resizes the resolution of the display adapater */ int err = 0;
- if (vc->vc_mode != KD_GRAPHICS && vc->vc_sw->con_resize) + if (vc->vc_sw->con_resize) err = vc->vc_sw->con_resize(vc, width, height, user);
return err; diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c index 3406067985b1..22bb3892f6bd 100644 --- a/drivers/video/fbdev/core/fbcon.c +++ b/drivers/video/fbdev/core/fbcon.c @@ -2019,7 +2019,7 @@ static int fbcon_resize(struct vc_data *vc, unsigned int width, return -EINVAL;
pr_debug("resize now %ix%i\n", var.xres, var.yres); - if (con_is_visible(vc)) { + if (con_is_visible(vc) && vc->vc_mode == KD_TEXT) { var.activate = FB_ACTIVATE_NOW | FB_ACTIVATE_FORCE; fb_set_var(info, &var);
On Sat, 15 May 2021, Tetsuo Handa wrote:
diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c index 3406067985b1..22bb3892f6bd 100644 --- a/drivers/video/fbdev/core/fbcon.c +++ b/drivers/video/fbdev/core/fbcon.c @@ -2019,7 +2019,7 @@ static int fbcon_resize(struct vc_data *vc, unsigned int width, return -EINVAL;
pr_debug("resize now %ix%i\n", var.xres, var.yres);
if (con_is_visible(vc)) {
if (con_is_visible(vc) && vc->vc_mode == KD_TEXT) { var.activate = FB_ACTIVATE_NOW | FB_ACTIVATE_FORCE; fb_set_var(info, &var);
LGTM, although I'll yet try to verify it with hardware. But it'll have to wait another week or so as I'm currently away from my lab and this requires physical presence.
Reviewed-by: Maciej W. Rozycki macro@orcam.me.uk
Maciej
On Sat, 15 May 2021, Maciej W. Rozycki wrote:
On Sat, 15 May 2021, Tetsuo Handa wrote:
diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c index 3406067985b1..22bb3892f6bd 100644 --- a/drivers/video/fbdev/core/fbcon.c +++ b/drivers/video/fbdev/core/fbcon.c @@ -2019,7 +2019,7 @@ static int fbcon_resize(struct vc_data *vc, unsigned int width, return -EINVAL;
pr_debug("resize now %ix%i\n", var.xres, var.yres);
if (con_is_visible(vc)) {
if (con_is_visible(vc) && vc->vc_mode == KD_TEXT) { var.activate = FB_ACTIVATE_NOW | FB_ACTIVATE_FORCE; fb_set_var(info, &var);LGTM, although I'll yet try to verify it with hardware. But it'll have to wait another week or so as I'm currently away from my lab and this requires physical presence.
NB I suggest that you request your change to be backported, i.e. post v3 with:
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org # v2.6.12+
Maciej
On Sat, May 15, 2021 at 9:33 AM Maciej W. Rozycki macro@orcam.me.uk wrote:
NB I suggest that you request your change to be backported, i.e. post v3 with:
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org # v2.6.12+
I've applied it to my tree, but let's wait to see that it doesn't cause any issues before notifying the stable people.
Linus
On Sat, May 15, 2021 at 6:42 PM Linus Torvalds torvalds@linux-foundation.org wrote:
On Sat, May 15, 2021 at 9:33 AM Maciej W. Rozycki macro@orcam.me.uk wrote:
NB I suggest that you request your change to be backported, i.e. post v3 with:
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org # v2.6.12+
I've applied it to my tree, but let's wait to see that it doesn't cause any issues before notifying the stable people.
Ah I missed all the fun with the long w/e. fwiw I think this looks very reasonable, see my other reply why I think this shouldn't cause issues. Especially when fbcon_resize only touches hw when in KD_TEXT mode. -Daniel
On Fri, 14 May 2021, Linus Torvalds wrote:
Overall I think it does make sense to resize the text console at any time, even if the visible console (VT) chosen is in the graphics mode,
It might make sense, but only if we call the function to update the low-level data.
Not calling it, and then starting to randomly use the (wrong) geometry, and just limiting it so that it's all within the buffer - THAT does not make sense.
So I think your patch is fundamentally wrong. It basically says "let's use random stale incorrect data, but just make sure that the end result is still within the allocated buffer".
I guess you mean Tetsuo-san's patch, right? I haven't sent any in this discussion.
My patch is at least conceptually sane.
An alternative would be to just remove the "vcmode != KD_GRAPHICS" check entirely, and always call con_resize() to update the low-level data, but honestly, that seems very likelty to break something very fundamentally, since it's not how any of fbcon has ever been tested,
Umm, there isn't much to change as far as console data structures are concerned with a resize: obviously the width and the height, which affect the size of the character/attribute buffer, and maybe some cursor data such as the size and screen coordinates.
For vgacon we have:
if (con_is_visible(c) && !vga_is_gfx) /* who knows */ vgacon_doresize(c, width, height);
in `vgacon_resize' already, following all the sanity checks, so the CRTC isn't poked at if `vga_is_gfx', exactly as we want.
I can see fbcon does not have equivalent code and instead has relied on the KD_GRAPHICS check made by the caller. Which I think has been a bug since fbcon's inception. Instead I think `fbcon_resize' ought to make all the sanity checks I can see it does and only then check for KD_GRAPHICS and if so, then exit without poking at hardware. Then upon exit from the gfx mode the `fb_set_var' call made from `fbcon_blank' will DTRT.
I can try verifying the latter hypothesis, though my framebuffer setups (with DECstation hardware) have always been somewhat incomplete. I do believe I have a MIPS fbdev X server binary somewhere to fiddle with, which should work with that TGA/SFB+ video adapter I mentioned before.
Another alternative would be to just delay the resize to when vcmode is put back to text mode again. That sounds somewhat reasonable to me, but it's a pretty big thing.
Methinks it works exactly like that already. On exit from the graphics mode (a VT switch or gfx program termination) hardware is reprogrammed according to the console geometry previously set. We just must not break it.
Maciej
On Fri, May 14, 2021 at 10:33 PM Linus Torvalds torvalds@linux-foundation.org wrote:
On Fri, May 14, 2021 at 1:25 PM Maciej W. Rozycki macro@orcam.me.uk wrote:
Overall I think it does make sense to resize the text console at any time, even if the visible console (VT) chosen is in the graphics mode,
It might make sense, but only if we call the function to update the low-level data.
Not calling it, and then starting to randomly use the (wrong) geometry, and just limiting it so that it's all within the buffer - THAT does not make sense.
So I think your patch is fundamentally wrong. It basically says "let's use random stale incorrect data, but just make sure that the end result is still within the allocated buffer".
My patch is at least conceptually sane.
An alternative would be to just remove the "vcmode != KD_GRAPHICS" check entirely, and always call con_resize() to update the low-level data, but honestly, that seems very likelty to break something very fundamentally, since it's not how any of fbcon has ever been tested,
Just an aside: I think with fbdev drivers this would go boom, because you'd have fbcon interferring with a direct /dev/fb/* user.
But if your fbdev driver is actually a drm modeset driver, then we have additional limitations: If the userspace accesses the display through /dev/dri/card0, then the kernel blocks all access through /dev/fb/* (including fbcon) to the actual display (it only goes into the buffer used for fbdev emulation). And everything would be fine.
Also generally you'd get away with this even in problematic cases, since usually you resize your console when looking at it, not when X or something else is using your fbdev direct access.
The one thing that's left out here a bit in the cold is userspace modeset drivers in X. Those would get hosed. But also, we stopped supporting those in at least i915/amd/radeon/nouveau drivers, automatically falling back to the fbdev stuff in most cases (with or without the drm drivers underneath that), and no one screamed. So probably not many users left.
So I /think/ we could wager this, if it's the least intrusive fix from the kernel pov. But it has some risks that we need to revert again if we break some of the really old use-cases here.
Cheers, Daniel
Another alternative would be to just delay the resize to when vcmode is put back to text mode again. That sounds somewhat reasonable to me, but it's a pretty big thing.
But no, your patch to just "knowingly use entirely wrong values, then add a limit check because we know the values are possibly garbage and not consistent with reality" is simply not acceptable.
Linus
On Mon, May 17, 2021 at 3:07 PM Daniel Vetter daniel@ffwll.ch wrote:
On Fri, May 14, 2021 at 10:33 PM Linus Torvalds torvalds@linux-foundation.org wrote:
On Fri, May 14, 2021 at 1:25 PM Maciej W. Rozycki macro@orcam.me.uk wrote:
Overall I think it does make sense to resize the text console at any time, even if the visible console (VT) chosen is in the graphics mode,
It might make sense, but only if we call the function to update the low-level data.
Not calling it, and then starting to randomly use the (wrong) geometry, and just limiting it so that it's all within the buffer - THAT does not make sense.
So I think your patch is fundamentally wrong. It basically says "let's use random stale incorrect data, but just make sure that the end result is still within the allocated buffer".
My patch is at least conceptually sane.
An alternative would be to just remove the "vcmode != KD_GRAPHICS" check entirely, and always call con_resize() to update the low-level data, but honestly, that seems very likelty to break something very fundamentally, since it's not how any of fbcon has ever been tested,
Just an aside: I think with fbdev drivers this would go boom, because you'd have fbcon interferring with a direct /dev/fb/* user.
Boom here means a bit of screen corruption, because fbcon overdraws your X sessions. Fixed by the next redraw of X.
But if your fbdev driver is actually a drm modeset driver, then we have additional limitations: If the userspace accesses the display through /dev/dri/card0, then the kernel blocks all access through /dev/fb/* (including fbcon) to the actual display (it only goes into the buffer used for fbdev emulation). And everything would be fine.
Also generally you'd get away with this even in problematic cases, since usually you resize your console when looking at it, not when X or something else is using your fbdev direct access.
The one thing that's left out here a bit in the cold is userspace modeset drivers in X. Those would get hosed. But also, we stopped supporting those in at least i915/amd/radeon/nouveau drivers, automatically falling back to the fbdev stuff in most cases (with or without the drm drivers underneath that), and no one screamed. So probably not many users left.
This one could lead to incosistent hw state, which would be worse.
So I /think/ we could wager this, if it's the least intrusive fix from the kernel pov. But it has some risks that we need to revert again if we break some of the really old use-cases here.
Cheers, Daniel
Another alternative would be to just delay the resize to when vcmode is put back to text mode again. That sounds somewhat reasonable to me, but it's a pretty big thing.
But no, your patch to just "knowingly use entirely wrong values, then add a limit check because we know the values are possibly garbage and not consistent with reality" is simply not acceptable.
Linus-- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch
On 2021/05/15 5:25, Maciej W. Rozycki wrote:
NB for fbcon the usual ioctl to resize the console is FBIOPUT_VSCREENINFO rather than VT_RESIZEX; fbset(8) uses it, and I actually experimented with it and a TGA-like (SFB+) framebuffer when at my lab last time, as Linux is kind enough to know how to fiddle with its clockchip. It works just fine.
fbcon_update_vcs() from FBIOPUT_VSCREENINFO is no-op if vc->vc_mode != KD_TEXT (which is equivalent to "if vc->vc_mode == KD_GRAPHICS" because KD_TEXT0/KD_TEXT1 are treated as KD_TEXT). Then, maybe it is OK to let resize_screen() return -EINVAL in order to make vc_do_resize() request fail if vc->vc_mode == KD_GRAPHICS.
Overall I think it does make sense to resize the text console at any time, even if the visible console (VT) chosen is in the graphics mode, as my understanding (and experience at least with vgacon) is that resizing the console applies globally across all the VTs. So the intent of the original change appears valid to me, and the choice not to reprogram the visible console and only store the settings for a future use if it's in the graphics mode correct.
Which means any bug triggered here needs to be fixed elsewhere rather than by making the request fail.
Since syzbot does not trigger this problem with Linus's patch, I think we can try Linus's patch with
pr_info_once("Resizing text console while in graphical mode is ignored. Please report if you need this.\n");
added in order to see if somebody wants "only store the settings for a future use".
dri-devel@lists.freedesktop.org
-
Daniel Vetter -
Linus Torvalds -
Maciej W. Rozycki -
syzbot -
Tetsuo Handa