Re: RFC: Comments on patches for libdrm are welcomed

[Sergey Bugaev]

On Sun, Oct 31, 2021 at 8:44 PM Svante Signell <svante.signell@gmail.com> wrote:
> One alternative to the path_max.diff patch is to simply do:
> #ifndef PATH_MAX
> #define PATH_MAX 4096
> #endif
> in xf86drm.c. This is a patch that upstream would accept more easily than the
> proposed one, but is not very Hurdish.
>
> WDYT?
>
> Thanks!

I see libdrm already uses asprintf and vasprintf in a few places, and
the developers seem to be fine with more usage of it [0], so it should
be possible to avoid the little mess of calling snprintf twice and an
explicit malloc.

[0]: https://gitlab.freedesktop.org/mesa/drm/-/merge_requests/108#note_790401

I.e.,

--- libdrm-2.4.107.orig/xf86drm.c
+++ libdrm-2.4.107/xf86drm.c
@@ -3359,7 +3359,8 @@ static char *drmGetMinorNameForFD(int fd
     return strdup(name);
 #else
     struct stat sbuf;
-    char buf[PATH_MAX + 1];
+    char *buf;
     const char *dev_name = drmGetDeviceName(type);
     unsigned int maj, min;
     int n;
@@ -3376,11 +3377,19 @@ static char *drmGetMinorNameForFD(int fd
     if (!dev_name)
         return NULL;

-    n = snprintf(buf, sizeof(buf), dev_name, DRM_DIR_NAME, min);
+    n = asprintf(&buf, dev_name, DRM_DIR_NAME, min);
-    if (n == -1 || n >= sizeof(buf))
+    if (n == -1)
         return NULL;

-    return strdup(buf);
+    return buf;
 #endif
 }

Sergey