Re: [Intel-gfx] [PATCH 1/6] igt_kms: don't leak the mmap used for cairo

2014-01-02 Thread Paulo Zanoni
2013/12/31 Chris Wilson : > On Mon, Dec 30, 2013 at 06:16:20PM +, Chris Wilson wrote: >> On Mon, Dec 30, 2013 at 03:53:10PM -0200, Paulo Zanoni wrote: >> > 2013/12/30 Chris Wilson : >> > > On Mon, Dec 30, 2013 at 01:56:48PM -0200, Paulo Zanoni wrote: >> > >> From: Paulo Zanoni >> > >> >> > >>

Re: [Intel-gfx] [PATCH 1/6] igt_kms: don't leak the mmap used for cairo

2013-12-31 Thread Chris Wilson
On Mon, Dec 30, 2013 at 06:16:20PM +, Chris Wilson wrote: > On Mon, Dec 30, 2013 at 03:53:10PM -0200, Paulo Zanoni wrote: > > 2013/12/30 Chris Wilson : > > > On Mon, Dec 30, 2013 at 01:56:48PM -0200, Paulo Zanoni wrote: > > >> From: Paulo Zanoni > > >> > > >> When we call kmstest_get_cairo_ctx

Re: [Intel-gfx] [PATCH 1/6] igt_kms: don't leak the mmap used for cairo

2013-12-30 Thread Chris Wilson
On Mon, Dec 30, 2013 at 03:53:10PM -0200, Paulo Zanoni wrote: > 2013/12/30 Chris Wilson : > > On Mon, Dec 30, 2013 at 01:56:48PM -0200, Paulo Zanoni wrote: > >> From: Paulo Zanoni > >> > >> When we call kmstest_get_cairo_ctx() and create a context, we do a > >> gem_mmap. The problem is that we los

Re: [Intel-gfx] [PATCH 1/6] igt_kms: don't leak the mmap used for cairo

2013-12-30 Thread Paulo Zanoni
2013/12/30 Chris Wilson : > On Mon, Dec 30, 2013 at 01:56:48PM -0200, Paulo Zanoni wrote: >> From: Paulo Zanoni >> >> When we call kmstest_get_cairo_ctx() and create a context, we do a >> gem_mmap. The problem is that we lose the mmap pointer, so we leak it. >> This patch stores the pointer and fr

Re: [Intel-gfx] [PATCH 1/6] igt_kms: don't leak the mmap used for cairo

2013-12-30 Thread Chris Wilson
On Mon, Dec 30, 2013 at 01:56:48PM -0200, Paulo Zanoni wrote: > From: Paulo Zanoni > > When we call kmstest_get_cairo_ctx() and create a context, we do a > gem_mmap. The problem is that we lose the mmap pointer, so we leak it. > This patch stores the pointer and frees it at kmstest_remove_fb. >

[Intel-gfx] [PATCH 1/6] igt_kms: don't leak the mmap used for cairo

2013-12-30 Thread Paulo Zanoni
From: Paulo Zanoni When we call kmstest_get_cairo_ctx() and create a context, we do a gem_mmap. The problem is that we lose the mmap pointer, so we leak it. This patch stores the pointer and frees it at kmstest_remove_fb. Huge test suites like kms_flip do this operation thousands of times, which