tests/gem_error_capture: Initial testcase for error state capture/dump
PATCH 1/1 tests/gem_error_capture: Initial testcase for error state capture/dump
From: Oscar Mateo <[email protected]>
--- tests/.gitignore | 1 + tests/Makefile.sources | 1 + tests/gem_error_capture.c | 230 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 232 insertions(+) create mode 100644 tests/gem_error_capture.c
tests/.gitignore +1 -0
@@ -27,6 +27,7 @@ gem_ctx_create gem_ctx_exec gem_double_irq_loop gem_dummy_reloc_loop +gem_error_capture gem_evict_alignment gem_evict_everything gem_exec_bad_domains
tests/Makefile.sources +1 -0
@@ -24,6 +24,7 @@ TESTS_progs_M = \ gem_ctx_bad_exec \ gem_ctx_exec \ gem_dummy_reloc_loop \ + gem_error_capture \ gem_evict_alignment \ gem_evict_everything \ gem_exec_bad_domains \
tests/gem_error_capture.c new file +230 -0
230 lines not on this page · raw message
Signature · 6 lines
1.9.0 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Daniel, is this what you had in mind?
-- Oscar
P.S. I just re-read the Jira task and realized that I am missing the "check that the ring objects contains a reloc with MI_BB_START for your presumed batch object's address". I´ll add this and resubmit.
Quoted 294 lines
-----Original Message-----
From: Mateo Lozano, Oscar
Sent: Friday, April 11, 2014 2:00 PM
To: [email protected]
Cc: Mateo Lozano, Oscar
Subject: [PATCH] tests/gem_error_capture: Initial testcase for error state capture/dumpFrom: Oscar Mateo <[email protected]>
Signed-off-by Oscar Mateo <[email protected]>--- tests/.gitignore | 1 + tests/Makefile.sources | 1 + tests/gem_error_capture.c | 230 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 232 insertions(+) create mode 100644 tests/gem_error_capture.ctests/.gitignore +1
-0@@ -27,6 +27,7 @@ gem_ctx_create gem_ctx_exec gem_double_irq_loop gem_dummy_reloc_loop +gem_error_capture gem_evict_alignment gem_evict_everything gem_exec_bad_domainstests/Makefile.sources index renamed from tests/Makefile.sources +0
-0bf02a48..612beb6 100644
tests/Makefile.sources +1
-0@@ -24,6 +24,7 @@ TESTS_progs_M = \ gem_ctx_bad_exec \ gem_ctx_exec \ gem_dummy_reloc_loop \ + gem_error_capture \ gem_evict_alignment \ gem_evict_everything \ gem_exec_bad_domains \tests/gem_error_capture.c new file renamed from tests/gem_error_capture.c +0
-0mode 100644 index 0000000..bbf0f5d
tests/gem_error_capture.c new file +26
-0@@ -0,0 +1,230 @@ +/* +* Copyright © 2014 Intel Corporation +* +* Permission is hereby granted, free of charge, to any person +obtaining a +* copy of this software and associated documentation files (the +"Software"), +* to deal in the Software without restriction, including without +limitation +* the rights to use, copy, modify, merge, publish, distribute, +sublicense, +* and/or sell copies of the Software, and to permit persons to whom +the +* Software is furnished to do so, subject to the following conditions: +* +* The above copyright notice and this permission notice (including the +next +* paragraph) shall be included in all copies or substantial portions +of the +* Software. +* +* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, +EXPRESS OR +* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, +* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
+SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES ORtests/gem_error_capture.c new file +4
-0+OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OROTHER
tests/gem_error_capture.c new file +115
-0+DEALINGS +* IN THE SOFTWARE. +* +* Authors: +* Oscar Mateo <[email protected]> +* +*/ + +/* +* Testcase: Check whether basic error state capture/dump mechanism is +correctly +* working for all rings. +*/ + +#include <unistd.h> +#include <stdlib.h> +#include <stdint.h> +#include <stdio.h> +#include <string.h> +#include <fcntl.h> +#include <inttypes.h> +#include <errno.h> +#include <sys/stat.h> +#include <sys/ioctl.h> +#include "drm.h" +#include "ioctl_wrappers.h" +#include "drmtest.h" +#include "intel_io.h" +#include "igt_debugfs.h" + +#define MAGIC_NUMBER 0x10001 +uint32_t batch[4 ] = {MI_NOOP ,MI_BATCH_BUFFER_END ,MAGIC_NUMBER , +MAGIC_NUMBER }; + +static void stop_rings (void ) +{ +int fd; +static const char buf[] ="0xf" ; + + fd =igt_debugfs_open ("i915_ring_stop" ,O_WRONLY ); +igt_assert (fd >=0 ); + +igt_assert (write (fd, buf,sizeof (buf)) ==sizeof (buf)); + +close (fd); +} + +static bool rings_stopped (void ) +{ +int fd; +static char buf[128 ]; +unsigned long long val; + + fd =igt_debugfs_open ("i915_ring_stop" ,O_RDONLY ); +igt_assert (fd >=0 ); + +igt_assert (read (fd, buf,sizeof (buf)) >0 ); +close (fd); + +sscanf (buf,"%llx" , &val); + +return (bool )val; +} + +static void clear_error_state (void ) +{ +int fd; +static const char buf[] ="" ; + + fd =igt_debugfs_open ("i915_error_state" ,O_WRONLY ); +igt_assert (fd >=0 ); + +igt_assert (write (fd, buf,sizeof (buf)) ==sizeof (buf)); +close (fd); +} + +static void check_bb_contents (char **line,size_t *line_size, +FILE *file,int pos,uint32_t expected_value) { +char expected_line[32 ]; + +igt_assert (getline (line, line_size, file) >0 ); + +snprintf (expected_line,sizeof (expected_line),"%08x : %08x" , +4 *pos, expected_value); + +igt_assert (strstr (*line, expected_line)); } + +static void check_error_state (const char *expected_ring_name, +uint64_t expected_offset) +{ +FILE *file; +int debug_fd; +char *line =NULL ; +size_t line_size; +char *dashes =NULL ; +char *ring_name =NULL ; +uint32_t gtt_offset =0 ; +bool matched; +int i; + + debug_fd =igt_debugfs_open ("i915_error_state" ,O_RDONLY ); +igt_assert (debug_fd >=0 ); + file =fdopen (debug_fd,"r" ); + +while (getline (&line, &line_size, file) >0 ) { + dashes =strstr (line,"---" ); +if (dashes) { + ring_name =realloc (ring_name, dashes - line); +strncpy (ring_name, line, dashes - line); + ring_name[dashes - line -1 ] ='\0' ; + + matched =sscanf (dashes,"--- gtt_offset = 0x%08x\n" , + >t_offset); +if (matched) { + igt_assert(strstr(ring_name,
expected_ring_name));tests/gem_error_capture.c new file +95
-1+igt_assert (gtt_offset == expected_offset); + +for (i =0 ; i <sizeof (batch) /4 ; i++) +check_bb_contents (&line, &line_size, + file, i, batch[i]); + +break ; + } + } + } + +igt_assert (matched); + +free (line); +free (ring_name); + +close (debug_fd); +} + +static void test (int fd,uint32_t handle,unsigned ring_id,const char +*ring_name) { +struct drm_i915_gem_execbuffer2 execbuf; +struct drm_i915_gem_exec_object2 exec; +uint64_t presumed_offset; + +gem_require_ring (fd, ring_id); + +clear_error_state (); + + exec.handle = handle; + exec.relocation_count =0 ; + exec.relocs_ptr =0 ; + exec.alignment =0 ; + exec.offset =0 ; + exec.flags =0 ; + exec.rsvd1 =0 ; + exec.rsvd2 =0 ; + + execbuf.buffers_ptr = (uintptr_t )&exec; + execbuf.buffer_count =1 ; + execbuf.batch_start_offset =0 ; + execbuf.batch_len =16 ; + execbuf.cliprects_ptr =0 ; + execbuf.num_cliprects =0 ; + execbuf.DR1 =0 ; + execbuf.DR4 =0 ; + execbuf.flags = ring_id; +i915_execbuffer2_set_context_id (execbuf,0 ); + execbuf.rsvd2 =0 ; + +gem_execbuf (fd, &execbuf); +gem_sync (fd, handle); + + presumed_offset = exec.offset; + +stop_rings (); + +gem_execbuf (fd, &execbuf); +gem_sync (fd, handle); + +igt_assert (rings_stopped () == false); +igt_assert (presumed_offset == exec.offset); + +check_error_state (ring_name, exec.offset); } + +uint32_t handle; +int fd; + +igt_main +{ + igt_fixture { + fd =drm_open_any (); + + handle =gem_create (fd,4096 ); +gem_write (fd, handle,0 , batch,sizeof (batch)); + } + + igt_subtest("render") +test (fd, handle,I915_EXEC_RENDER ,"render ring" ); + + igt_subtest("bsd") +test (fd, handle,I915_EXEC_BSD ,"bsd ring" ); + + igt_subtest("blt") +test (fd, handle,I915_EXEC_BLT ,"blitter ring" ); + + igt_subtest("vebox") +test (fd, handle,I915_EXEC_VEBOX ,"video enhancement ring" ); + + igt_fixture { +gem_close (fd, handle); + +close (fd); + } +} --
Quoted 1 line
1.9.0
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 7 lines · Oscar
Daniel, is this what you had in mind?
-- Oscar
P.S. I just re-read the Jira task and realized that I am missing the "check that the ring objects contains a reloc withMI_BB_STARTfor your presumed batch object's address". I´ll add this and resubmit.
Yeah this looks pretty cool. btw when you resend can you simply sign up someone local to your team for the review? Bit easier that way.
I've checked your usage of igt_* infrastructure and that looks good.
Thanks, Daniel
Quoted 297 lines
Quoted 296 lines
-----Original Message-----
From: Mateo Lozano, Oscar
Sent: Friday, April 11, 2014 2:00 PM
To: [email protected]
Cc: Mateo Lozano, Oscar
Subject: [PATCH] tests/gem_error_capture: Initial testcase for error state capture/dumpFrom: Oscar Mateo <[email protected]>
Signed-off-by Oscar Mateo <[email protected]>--- tests/.gitignore | 1 + tests/Makefile.sources | 1 + tests/gem_error_capture.c | 230 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 232 insertions(+) create mode 100644 tests/gem_error_capture.ctests/.gitignore +1
-0@@ -27,6 +27,7 @@ gem_ctx_create gem_ctx_exec gem_double_irq_loop gem_dummy_reloc_loop +gem_error_capture gem_evict_alignment gem_evict_everything gem_exec_bad_domainstests/Makefile.sources index renamed from tests/Makefile.sources +0
-0bf02a48..612beb6 100644
tests/Makefile.sources +1
-0@@ -24,6 +24,7 @@ TESTS_progs_M = \ gem_ctx_bad_exec \ gem_ctx_exec \ gem_dummy_reloc_loop \ + gem_error_capture \ gem_evict_alignment \ gem_evict_everything \ gem_exec_bad_domains \tests/gem_error_capture.c new file renamed from tests/gem_error_capture.c +0
-0mode 100644 index 0000000..bbf0f5d
tests/gem_error_capture.c new file +26
-0@@ -0,0 +1,230 @@ +/* +* Copyright © 2014 Intel Corporation +* +* Permission is hereby granted, free of charge, to any person +obtaining a +* copy of this software and associated documentation files (the +"Software"), +* to deal in the Software without restriction, including without +limitation +* the rights to use, copy, modify, merge, publish, distribute, +sublicense, +* and/or sell copies of the Software, and to permit persons to whom +the +* Software is furnished to do so, subject to the following conditions: +* +* The above copyright notice and this permission notice (including the +next +* paragraph) shall be included in all copies or substantial portions +of the +* Software. +* +* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, +EXPRESS OR +* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, +* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
+SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES ORtests/gem_error_capture.c new file +4
-0+OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OROTHER
tests/gem_error_capture.c new file +115
-0+DEALINGS +* IN THE SOFTWARE. +* +* Authors: +* Oscar Mateo <[email protected]> +* +*/ + +/* +* Testcase: Check whether basic error state capture/dump mechanism is +correctly +* working for all rings. +*/ + +#include <unistd.h> +#include <stdlib.h> +#include <stdint.h> +#include <stdio.h> +#include <string.h> +#include <fcntl.h> +#include <inttypes.h> +#include <errno.h> +#include <sys/stat.h> +#include <sys/ioctl.h> +#include "drm.h" +#include "ioctl_wrappers.h" +#include "drmtest.h" +#include "intel_io.h" +#include "igt_debugfs.h" + +#define MAGIC_NUMBER 0x10001 +uint32_t batch[4 ] = {MI_NOOP ,MI_BATCH_BUFFER_END ,MAGIC_NUMBER , +MAGIC_NUMBER }; + +static void stop_rings (void ) +{ +int fd; +static const char buf[] ="0xf" ; + + fd =igt_debugfs_open ("i915_ring_stop" ,O_WRONLY ); +igt_assert (fd >=0 ); + +igt_assert (write (fd, buf,sizeof (buf)) ==sizeof (buf)); + +close (fd); +} + +static bool rings_stopped (void ) +{ +int fd; +static char buf[128 ]; +unsigned long long val; + + fd =igt_debugfs_open ("i915_ring_stop" ,O_RDONLY ); +igt_assert (fd >=0 ); + +igt_assert (read (fd, buf,sizeof (buf)) >0 ); +close (fd); + +sscanf (buf,"%llx" , &val); + +return (bool )val; +} + +static void clear_error_state (void ) +{ +int fd; +static const char buf[] ="" ; + + fd =igt_debugfs_open ("i915_error_state" ,O_WRONLY ); +igt_assert (fd >=0 ); + +igt_assert (write (fd, buf,sizeof (buf)) ==sizeof (buf)); +close (fd); +} + +static void check_bb_contents (char **line,size_t *line_size, +FILE *file,int pos,uint32_t expected_value) { +char expected_line[32 ]; + +igt_assert (getline (line, line_size, file) >0 ); + +snprintf (expected_line,sizeof (expected_line),"%08x : %08x" , +4 *pos, expected_value); + +igt_assert (strstr (*line, expected_line)); } + +static void check_error_state (const char *expected_ring_name, +uint64_t expected_offset) +{ +FILE *file; +int debug_fd; +char *line =NULL ; +size_t line_size; +char *dashes =NULL ; +char *ring_name =NULL ; +uint32_t gtt_offset =0 ; +bool matched; +int i; + + debug_fd =igt_debugfs_open ("i915_error_state" ,O_RDONLY ); +igt_assert (debug_fd >=0 ); + file =fdopen (debug_fd,"r" ); + +while (getline (&line, &line_size, file) >0 ) { + dashes =strstr (line,"---" ); +if (dashes) { + ring_name =realloc (ring_name, dashes - line); +strncpy (ring_name, line, dashes - line); + ring_name[dashes - line -1 ] ='\0' ; + + matched =sscanf (dashes,"--- gtt_offset = 0x%08x\n" , + >t_offset); +if (matched) { + igt_assert(strstr(ring_name,
expected_ring_name));tests/gem_error_capture.c new file +95
-1+igt_assert (gtt_offset == expected_offset); + +for (i =0 ; i <sizeof (batch) /4 ; i++) +check_bb_contents (&line, &line_size, + file, i, batch[i]); + +break ; + } + } + } + +igt_assert (matched); + +free (line); +free (ring_name); + +close (debug_fd); +} + +static void test (int fd,uint32_t handle,unsigned ring_id,const char +*ring_name) { +struct drm_i915_gem_execbuffer2 execbuf; +struct drm_i915_gem_exec_object2 exec; +uint64_t presumed_offset; + +gem_require_ring (fd, ring_id); + +clear_error_state (); + + exec.handle = handle; + exec.relocation_count =0 ; + exec.relocs_ptr =0 ; + exec.alignment =0 ; + exec.offset =0 ; + exec.flags =0 ; + exec.rsvd1 =0 ; + exec.rsvd2 =0 ; + + execbuf.buffers_ptr = (uintptr_t )&exec; + execbuf.buffer_count =1 ; + execbuf.batch_start_offset =0 ; + execbuf.batch_len =16 ; + execbuf.cliprects_ptr =0 ; + execbuf.num_cliprects =0 ; + execbuf.DR1 =0 ; + execbuf.DR4 =0 ; + execbuf.flags = ring_id; +i915_execbuffer2_set_context_id (execbuf,0 ); + execbuf.rsvd2 =0 ; + +gem_execbuf (fd, &execbuf); +gem_sync (fd, handle); + + presumed_offset = exec.offset; + +stop_rings (); + +gem_execbuf (fd, &execbuf); +gem_sync (fd, handle); + +igt_assert (rings_stopped () == false); +igt_assert (presumed_offset == exec.offset); + +check_error_state (ring_name, exec.offset); } + +uint32_t handle; +int fd; + +igt_main +{ + igt_fixture { + fd =drm_open_any (); + + handle =gem_create (fd,4096 ); +gem_write (fd, handle,0 , batch,sizeof (batch)); + } + + igt_subtest("render") +test (fd, handle,I915_EXEC_RENDER ,"render ring" ); + + igt_subtest("bsd") +test (fd, handle,I915_EXEC_BSD ,"bsd ring" ); + + igt_subtest("blt") +test (fd, handle,I915_EXEC_BLT ,"blitter ring" ); + + igt_subtest("vebox") +test (fd, handle,I915_EXEC_VEBOX ,"video enhancement ring" ); + + igt_fixture { +gem_close (fd, handle); + +close (fd); + } +} --1.9.0
Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
PATCH v2 1/1 tests/gem_error_capture: Initial testcase for error state capture/dump
In reply to [email protected]From: Oscar Mateo <[email protected]>
Guarantees that error capture works at a very basic level.
v2: Also check that the ring object contains a reloc with MI_BB_START for the presumed batch object's address.
--- tests/.gitignore | 1 + tests/Makefile.sources | 1 + tests/gem_error_capture.c | 269 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 271 insertions(+) create mode 100644 tests/gem_error_capture.c
tests/.gitignore +1 -0
@@ -27,6 +27,7 @@ gem_ctx_create gem_ctx_exec gem_double_irq_loop gem_dummy_reloc_loop +gem_error_capture gem_evict_alignment gem_evict_everything gem_exec_bad_domains
tests/Makefile.sources +1 -0
@@ -24,6 +24,7 @@ TESTS_progs_M = \ gem_ctx_bad_exec \ gem_ctx_exec \ gem_dummy_reloc_loop \ + gem_error_capture \ gem_evict_alignment \ gem_evict_everything \ gem_exec_bad_domains \
tests/gem_error_capture.c new file +269 -0
269 lines not on this page · raw message
Signature · 6 lines
1.9.0 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 10 lines · On Fri, Apr 11, 2014 at 05:48:12PM +0100, [email protected] wrote:
tests/gem_error_capture.c new file +10
-0+static void check_error_state (const char *expected_ring_name, +uint64_t expected_offset) +{ +FILE *file; +int debug_fd; +char *line =NULL ; +size_t line_size =0 ; +char *dashes =NULL ; +char *ring_name =NULL ; +int bb_matched =0 ;
tests/gem_error_capture.c line 113
Quoted 10 lines
tests/gem_error_capture.c new file +10
-0+uint32_t gtt_offset; +char expected_line[32 ]; +int req_matched =0 ; +int requests; +uint32_t seqno, tail; +long jiffies; +int ringbuf_matched =0 ; +unsigned int offset, command, expected_addr =0 ; +int i, items; +bool bb_ok = false, req_ok = false, ringbuf_ok = false;
Most of these are only of use in local scope.
Quoted 24 lines
tests/gem_error_capture.c new file +24
-0+ debug_fd =igt_debugfs_open ("i915_error_state" ,O_RDONLY ); +igt_assert (debug_fd >=0 ); + file =fdopen (debug_fd,"r" ); + +while (getline (&line, &line_size, file) >0 ) { + dashes =strstr (line,"---" ); +if (!dashes) +continue ; + + ring_name =realloc (ring_name, dashes - line); +strncpy (ring_name, line, dashes - line); + ring_name[dashes - line -1 ] ='\0' ; + + bb_matched =sscanf (dashes,"--- gtt_offset = 0x%08x\n" , + >t_offset); +if (bb_matched ==1 ) { +igt_assert (strstr (ring_name, expected_ring_name)); +igt_assert (gtt_offset == expected_offset); + +for (i =0 ; i <sizeof (batch) /4 ; i++) { +igt_assert (getline (&line, &line_size, file) >0 ); +snprintf (expected_line,sizeof (expected_line),"%08x : %08x" , +4 *i, batch[i]); +igt_assert (strstr (line, expected_line));
tests/gem_error_capture.c line 148
Quoted 10 lines
tests/gem_error_capture.c new file +10
-0+ } + bb_ok = true; +continue ; + } + + req_matched =sscanf (dashes,"--- %d requests\n" , + &requests); +if (req_matched ==1 ) { +igt_assert (strstr (ring_name, expected_ring_name)); +igt_assert (requests ==1 );
Bad assumption. You could still have the request from gem_quiescent_gpu() which may not have been retired before the error triggers.
tests/gem_error_capture.c line 153
Quoted 5 lines
tests/gem_error_capture.c new file +5
-0+ +igt_assert (getline (&line, &line_size, file) >0 ); + items =sscanf (line," seqno 0x%08x, emitted %ld, tail 0x%08x\n" , + &seqno, &jiffies, &tail); +igt_assert (items ==3 );
Bad. I may change the format. s/may/will/
Quoted 15 lines
tests/gem_error_capture.c new file +15
-0+ req_ok = true; +continue ; + } + + ringbuf_matched =sscanf (dashes,"--- ringbuffer = 0x%08x\n" , + >t_offset); +if (ringbuf_matched ==1 ) { +if (!strstr (ring_name, expected_ring_name)) +continue ; +igt_assert (req_ok); + +for (i =0 ; i < tail /4 ; i++) { +igt_assert (getline (&line, &line_size, file) >0 ); + items =sscanf (line,"%08x : %08x\n" , + &offset, &command);
tests/gem_error_capture.c line 178
Quoted 10 lines
tests/gem_error_capture.c new file +10
-0+igt_assert (items ==2 ); +if ((command &0x1F800000 ) ==MI_BATCH_BUFFER_START ) { +igt_assert (getline (&line, &line_size, file) >0 ); + items =sscanf (line,"%08x : %08x\n" , + &offset, &expected_addr); +igt_assert (items ==2 ); + i++; + } + } +igt_assert (expected_addr == expected_offset);
Bad. Some gen encode flags into the BB start address.
Quoted 13 lines
tests/gem_error_capture.c new file +13
-0+ ringbuf_ok = true; +continue ; + } + +if (bb_ok && req_ok && ringbuf_ok) +break ; + } +igt_assert (bb_ok && req_ok && ringbuf_ok); + +free (line); +free (ring_name); +close (debug_fd); +}
Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre
Quoted 11 lines
Quoted 11 lines
tests/gem_error_capture.c new file +11
-0+FILE *file; +int debug_fd; +char *line =NULL ; +size_t line_size =0 ; +char *dashes =NULL ; +char *ring_name =NULL ; +int bb_matched =0 ; +uint32_t gtt_offset; +char expected_line[32 ]; +int req_matched =0 ; +int requests;
Quoted 8 lines
tests/gem_error_capture.c line 113
Quoted 6 lines
tests/gem_error_capture.c new file +6
-0+uint32_t seqno, tail; +long jiffies; +int ringbuf_matched =0 ; +unsigned int offset, command, expected_addr =0 ; +int i, items; +bool bb_ok = false, req_ok = false, ringbuf_ok = false;Most of these are only of use in local scope.
ACK, will change.
Quoted 23 lines
Quoted 12 lines
tests/gem_error_capture.c new file +12
-0+ debug_fd =igt_debugfs_open ("i915_error_state" ,O_RDONLY ); +igt_assert (debug_fd >=0 ); + file =fdopen (debug_fd,"r" ); + +while (getline (&line, &line_size, file) >0 ) { + dashes =strstr (line,"---" ); +if (!dashes) +continue ; + + ring_name =realloc (ring_name, dashes - line); +strncpy (ring_name, line, dashes - line); + ring_name[dashes - line -1 ] ='\0' ;tests/gem_error_capture.c
Quoted 10 lines
tests/gem_error_capture.c new file +10
-0+ + bb_matched =sscanf (dashes,"--- gtt_offset = 0x%08x\n" , + >t_offset); +if (bb_matched ==1 ) { +igt_assert (strstr (ring_name, expected_ring_name)); +igt_assert (gtt_offset == expected_offset); + +for (i =0 ; i <sizeof (batch) /4 ; i++) { +igt_assert (getline (&line, &line_size, file) >0 ); + snprintf(expected_line, sizeof(expected_line),"%08x : %08x",
Quoted 16 lines
tests/gem_error_capture.c line 148
Quoted 12 lines
tests/gem_error_capture.c new file +12
-0+ 4*i, batch[i]); +igt_assert (strstr (line, expected_line)); + } + bb_ok = true; +continue ; + } + + req_matched =sscanf (dashes,"--- %d requests\n" , + &requests); +if (req_matched ==1 ) { +igt_assert (strstr (ring_name, expected_ring_name)); +igt_assert (requests ==1 );Bad assumption. You could still have the request from
gem_quiescent_gpu()which may not have been retired before the error triggers.
Hmmmm... but I make a first valid gem_execbuf()+gem_sync() before the one that actually hangs. Shouldn´t that make sure that all previous requests have been retired?
Quoted 8 lines
tests/gem_error_capture.c
Quoted 3 lines
tests/gem_error_capture.c new file +3
-0+ +igt_assert (getline (&line, &line_size, file) >0 ); + items = sscanf(line, " seqno 0x%08x, emitted %ld, tail0x%08x\n",
tests/gem_error_capture.c line 153
Quoted 2 lines
tests/gem_error_capture.c new file +2
-0+ &seqno, &jiffies, &tail); +igt_assert (items ==3 );Bad. I may change the format. s/may/will/
I still need to make sure I got a valid tail, so I need to know the format and fail if don´t understand it. Or do you want me to use a different tail, maybe the ringbuffer one?
Quoted 23 lines
Quoted 7 lines
tests/gem_error_capture.c new file +7
-0+ req_ok = true; +continue ; + } + + ringbuf_matched =sscanf (dashes,"--- ringbuffer = 0x%08x\n" , + >t_offset); +if (ringbuf_matched ==1 ) {tests/gem_error_capture.c
Quoted 10 lines
tests/gem_error_capture.c new file +10
-0+if (!strstr (ring_name, expected_ring_name)) +continue ; +igt_assert (req_ok); + +for (i =0 ; i < tail /4 ; i++) { +igt_assert (getline (&line, &line_size, file) >0 ); + items =sscanf (line,"%08x : %08x\n" , + &offset, &command); +igt_assert (items ==2 ); + if ((command & 0x1F800000) ==
MI_BATCH_BUFFER_START) {Review
Quoted 1 line
(quoted diff) +1
-0+ igt_assert(getline(&line, &line_size,file) > 0);
Review
Quoted 2 lines
(quoted diff) +2
-0+ items = sscanf(line, "%08x : %08x\n", + &offset,
&expected_addr);
Quoted 7 lines
tests/gem_error_capture.c line 178
Quoted 5 lines
tests/gem_error_capture.c new file +5
-0+igt_assert (items ==2 ); + i++; + } + } +igt_assert (expected_addr == expected_offset);Bad. Some gen encode flags into the BB start address.
ACK, will change.
Quoted 13 lines
Quoted 13 lines
tests/gem_error_capture.c new file +13
-0+ ringbuf_ok = true; +continue ; + } + +if (bb_ok && req_ok && ringbuf_ok) +break ; + } +igt_assert (bb_ok && req_ok && ringbuf_ok); + +free (line); +free (ring_name); +close (debug_fd); +}
Quoted 10 lines · Oscar
Quoted 8 lines
tests/gem_error_capture.c line 148
Quoted 4 lines
tests/gem_error_capture.c new file +4
-0+ &requests); +if (req_matched ==1 ) { +igt_assert (strstr (ring_name, expected_ring_name)); +igt_assert (requests ==1 );Bad assumption. You could still have the request from
gem_quiescent_gpu()which may not have been retired before the error triggers.Hmmmm... but I make a first valid gem_execbuf()+gem_sync() before the one that actually hangs. Shouldn´t that make sure that all previous requests have been retired?
No, I would not make that assumption about kernel behaviour. The only thing that it will guarantee is that the last request with that handle is retired. (In other words, forthcoming bug fixes already break this assumption.)
Quoted 10 lines
Quoted 8 lines
tests/gem_error_capture.c
Quoted 3 lines
tests/gem_error_capture.c new file +3
-0+ +igt_assert (getline (&line, &line_size, file) >0 ); + items = sscanf(line, " seqno 0x%08x, emitted %ld, tail0x%08x\n",
tests/gem_error_capture.c line 153
Quoted 2 lines
tests/gem_error_capture.c new file +2
-0+ &seqno, &jiffies, &tail); +igt_assert (items ==3 );Bad. I may change the format. s/may/will/
I still need to make sure I got a valid tail, so I need to know the format and fail if don´t understand it. Or do you want me to use a different tail, maybe the ringbuffer one?
I didn't spot where you used tail. Care to elaborate? I may be able to suggest an alternative, or we may either code this more defensively or make the kernel emission easier to use and hopefully more informative for debugging.
-Chris
Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre
Quoted 9 lines
Quoted 9 lines
Quoted 8 lines
tests/gem_error_capture.c line 148
Quoted 4 lines
tests/gem_error_capture.c new file +4
-0+ &requests); +if (req_matched ==1 ) { +igt_assert (strstr (ring_name, expected_ring_name)); +igt_assert (requests ==1 );Bad assumption. You could still have the request from
gem_quiescent_gpu()which may not have been retired before the error triggers.
Quoted 8 lines
Quoted 1 line
Hmmmm... but I make a first valid gem_execbuf()+gem_sync() before the
one that actually hangs. Shouldn´t that make sure that all previous requests have been retired?
No, I would not make that assumption about kernel behaviour. The only thing that it will guarantee is that the last request with that handle is retired. (In other words, forthcoming bug fixes already break this
assumption.)
Ok, then I´ll check all the reported requests (can I at least assume that the one I am interested in will be the last one? request_list seems to be FIFO...).
Quoted 9 lines
Quoted 9 lines
Quoted 8 lines
tests/gem_error_capture.c
Quoted 3 lines
tests/gem_error_capture.c new file +3
-0+ +igt_assert (getline (&line, &line_size, file) >0 ); + items = sscanf(line, " seqno 0x%08x, emitted %ld, tail0x%08x\n",
tests/gem_error_capture.c line 153
Quoted 2 lines
tests/gem_error_capture.c new file +2
-0+ &seqno, &jiffies, &tail); +igt_assert (items ==3 );Bad. I may change the format. s/may/will/
Quoted 7 lines
Quoted 1 line
I still need to make sure I got a valid tail, so I need to know the format and
fail if don´t understand it. Or do you want me to use a different tail, maybe the ringbuffer one?
I didn't spot where you used tail. Care to elaborate? I may be able to suggest an alternative, or we may either code this more defensively or make the kernel emission easier to use and hopefully more informative for debugging.
I´m using the request tail to traverse the ring buffer, make sure there is a MI_BB_START and then our object´s address shortly before that tail. I could use ring->tail, or I could skip this check altogether (after all, I´m already checking that the GPU was executing a Batch Buffer, and that the BB had my magic number on it).
Quoted 13 lines · Oscar
Quoted 13 lines
Quoted 10 lines
Quoted 8 lines
tests/gem_error_capture.c line 148
Quoted 4 lines
tests/gem_error_capture.c new file +4
-0+ &requests); +if (req_matched ==1 ) { +igt_assert (strstr (ring_name, expected_ring_name)); +igt_assert (requests ==1 );Bad assumption. You could still have the request from
gem_quiescent_gpu()which may not have been retired before the error triggers.Hmmmm... but I make a first valid gem_execbuf()+gem_sync() before the
one that actually hangs. Shouldn´t that make sure that all previous requests have been retired?
Quoted 6 lines
Quoted 4 lines
No, I would not make that assumption about kernel behaviour. The only thing that it will guarantee is that the last request with that handle is retired. (In other words, forthcoming bug fixes already break this
assumption.)Ok, then I´ll check all the reported requests (can I at least assume that the one I am interested in will be the last one?
request_listseems to be FIFO...).
In general, even that may be dangerous. (But less so if can be sure that is no batchbuffer injection from the testsuite, and that there are no other clients, or that the kernel doesn't start having to do its own w/a.)
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 8 lines
tests/gem_error_capture.c
Quoted 3 lines
tests/gem_error_capture.c new file +3
-0+ +igt_assert (getline (&line, &line_size, file) >0 ); + items = sscanf(line, " seqno 0x%08x, emitted %ld, tail0x%08x\n",
tests/gem_error_capture.c line 153
Quoted 2 lines
tests/gem_error_capture.c new file +2
-0+ &seqno, &jiffies, &tail); +igt_assert (items ==3 );Bad. I may change the format. s/may/will/
I still need to make sure I got a valid tail, so I need to know the format and
Quoted 8 lines
Quoted 6 lines
fail if don´t understand it. Or do you want me to use a different tail, maybe the ringbuffer one?
I didn't spot where you used tail. Care to elaborate? I may be able to suggest an alternative, or we may either code this more defensively or make the kernel emission easier to use and hopefully more informative for debugging.
I´m using the request tail to traverse the ring buffer, make sure there is a
MI_BB_STARTand then our object´s address shortly before that tail. I could usering->tail, or I could skip this check altogether (after all, I´m already checking that the GPU was executing a Batch Buffer, and that the BB had my magic number on it).
I would add a little more smarts to both the kernel and error-decode. In the kernel, we can print the guilty request, which you can then use to confirm that it is yours. That seems to me to be a stronger validation of gem_error_capture, and a useful bit of information from hangstats that we do not expose currently.
-Chris
Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre
Quoted 5 lines
I would add a little more smarts to both the kernel and error-decode. In the kernel, we can print the guilty request, which you can then use to confirm that it is yours. That seems to me to be a stronger validation of
gem_error_capture, and a useful bit of information from hangstats that we do not expose currently.
That sounds good. I have to add a number of other things to i915_gpu_error as part of the Execlists code, so I´ll add a "--- guilty request" as well and resubmit this test together with the series.
Thanks,
Oscar
Quoted 9 lines · Oscar
Quoted 5 lines
I would add a little more smarts to both the kernel and error-decode. In the kernel, we can print the guilty request, which you can then use to confirm that it is yours. That seems to me to be a stronger validation of
gem_error_capture, and a useful bit of information from hangstats that we do not expose currently.That sounds good. I have to add a number of other things to
i915_gpu_erroras part of the Execlists code, so I´ll add a "--- guilty request" as well and resubmit this test together with the series.
If we want this much smarts then we need a properly hanging batch, e.g. like the looping batch used in gem_reset_stats.
The problem with that is that this will kill the gpu if reset doesn't work (i.e. gen2/3) so we need to skip this test there. Or maybe split things into 2 subtests and use the properly hanging batch only when we do the extended guilty testing under discussion here.
But in any case just checking that the batch is somewhere in the ring (properly masking of lower bits 0-11 ofc) and checking whether the batch is correctl dumped (with the magic value) would catch a lot of the past&present execbuf bugs - we've had issues with dumping fancy values of 0 a lot.
For the guilty stuff we have an extensive set of tests in gem_reset_stat using the reset stat ioctl already. And for the occasional "the hang detection logic is busted bug" I think nothing short of a human brain locking at the batch really helps. At least if we want to be somewhat platform agnostic ...
So imo the current level of checking loosk Good Enough. But I'm certainly not going to stop you ;-)
Cheers, Daniel
Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
Quoted 25 lines · Daniel Vetter
Quoted 9 lines · Oscar
Quoted 5 lines
I would add a little more smarts to both the kernel and error-decode. In the kernel, we can print the guilty request, which you can then use to confirm that it is yours. That seems to me to be a stronger validation of
gem_error_capture, and a useful bit of information from hangstats that we do not expose currently.That sounds good. I have to add a number of other things to
i915_gpu_erroras part of the Execlists code, so I´ll add a "--- guilty request" as well and resubmit this test together with the series.If we want this much smarts then we need a properly hanging batch, e.g. like the looping batch used in
gem_reset_stats.The problem with that is that this will kill the gpu if reset doesn't work (i.e. gen2/3) so we need to skip this test there. Or maybe split things into 2 subtests and use the properly hanging batch only when we do the extended guilty testing under discussion here.
But in any case just checking that the batch is somewhere in the ring (properly masking of lower bits 0-11 ofc) and checking whether the batch is correctl dumped (with the magic value) would catch a lot of the past&present execbuf bugs - we've had issues with dumping fancy values of 0 a lot.
Quoted 8 lines
For the guilty stuff we have an extensive set of tests in
gem_reset_statusing the reset stat ioctl already. And for the occasional "the hang detection logic is busted bug" I think nothing short of a human brain locking at the batch really helps. At least if we want to be somewhat platform agnostic ...So imo the current level of checking loosk Good Enough. But I'm certainly not going to stop you ;-)
Ok, for the moment I'm happy that I can unblock Execlists. I don't mind looking at the "--- guilty requests" in the future, but first I need to get Execlists out of the way...
Cheers,
Oscar
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Hi,
Quoted 115 lines · [email protected] writes:
From: Oscar Mateo <[email protected]>
Guarantees that error capture works at a very basic level.
v2: Also check that the ring object contains a reloc with
MI_BB_STARTfor the presumed batch object's address.Signed-off-by Oscar Mateo <[email protected]>--- tests/.gitignore | 1 + tests/Makefile.sources | 1 + tests/gem_error_capture.c | 269 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 271 insertions(+) create mode 100644 tests/gem_error_capture.ctests/.gitignore +1
-0@@ -27,6 +27,7 @@ gem_ctx_create gem_ctx_exec gem_double_irq_loop gem_dummy_reloc_loop +gem_error_capture gem_evict_alignment gem_evict_everything gem_exec_bad_domainstests/Makefile.sources +1
-0@@ -24,6 +24,7 @@ TESTS_progs_M = \ gem_ctx_bad_exec \ gem_ctx_exec \ gem_dummy_reloc_loop \ + gem_error_capture \ gem_evict_alignment \ gem_evict_everything \ gem_exec_bad_domains \tests/gem_error_capture.c new file +70
-0@@ -0,0 +1,269 @@ +/* +* Copyright © 2014 Intel Corporation +* +* Permission is hereby granted, free of charge, to any person obtaining a +* copy of this software and associated documentation files (the "Software"), +* to deal in the Software without restriction, including without limitation +* the rights to use, copy, modify, merge, publish, distribute, sublicense, +* and/or sell copies of the Software, and to permit persons to whom the +* Software is furnished to do so, subject to the following conditions: +* +* The above copyright notice and this permission notice (including the next +* paragraph) shall be included in all copies or substantial portions of the +* Software. +* +* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING +* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS +* IN THE SOFTWARE. +* +* Authors: +* Oscar Mateo <[email protected]> +* +*/ + +/* +* Testcase: Check whether basic error state capture/dump mechanism is correctly +* working for all rings. +*/ + +#include <unistd.h> +#include <stdlib.h> +#include <stdint.h> +#include <stdio.h> +#include <string.h> +#include <fcntl.h> +#include <inttypes.h> +#include <errno.h> +#include <sys/stat.h> +#include <sys/ioctl.h> +#include "drm.h" +#include "ioctl_wrappers.h" +#include "drmtest.h" +#include "intel_io.h" +#include "igt_debugfs.h" + +#define MAGIC_NUMBER 0x10001 +uint32_t batch[4 ] = {MI_NOOP ,MI_BATCH_BUFFER_END ,MAGIC_NUMBER ,MAGIC_NUMBER }; + +static void stop_rings (void ) +{ +int fd; +static const char buf[] ="0xf" ; + + fd =igt_debugfs_open ("i915_ring_stop" ,O_WRONLY ); +igt_assert (fd >=0 ); + +igt_assert (write (fd, buf,sizeof (buf)) ==sizeof (buf)); + +close (fd); +} + +static bool rings_stopped (void ) +{ +int fd; +static char buf[128 ]; +unsigned long long val; +
tests/gem_error_capture.c line 80
Quoted 10 lines
tests/gem_error_capture.c new file +10
-0+ fd =igt_debugfs_open ("i915_ring_stop" ,O_RDONLY ); +igt_assert (fd >=0 ); + +igt_assert (read (fd, buf,sizeof (buf)) >0 ); +close (fd); + +sscanf (buf,"%llx" , &val); + +return (bool )val; +}
Please use igt_set_stop_rings() and igt_get_stop_rings().
Also consider stopping only the ring you are testing for.
Quoted 83 lines
tests/gem_error_capture.c new file +83
-0+static void clear_error_state (void ) +{ +int fd; +static const char buf[] ="" ; + + fd =igt_debugfs_open ("i915_error_state" ,O_WRONLY ); +igt_assert (fd >=0 ); + +igt_assert (write (fd, buf,sizeof (buf)) ==sizeof (buf)); +close (fd); +} + +static void check_error_state (const char *expected_ring_name, +uint64_t expected_offset) +{ +FILE *file; +int debug_fd; +char *line =NULL ; +size_t line_size =0 ; +char *dashes =NULL ; +char *ring_name =NULL ; +int bb_matched =0 ; +uint32_t gtt_offset; +char expected_line[32 ]; +int req_matched =0 ; +int requests; +uint32_t seqno, tail; +long jiffies; +int ringbuf_matched =0 ; +unsigned int offset, command, expected_addr =0 ; +int i, items; +bool bb_ok = false, req_ok = false, ringbuf_ok = false; + + debug_fd =igt_debugfs_open ("i915_error_state" ,O_RDONLY ); +igt_assert (debug_fd >=0 ); + file =fdopen (debug_fd,"r" ); + +while (getline (&line, &line_size, file) >0 ) { + dashes =strstr (line,"---" ); +if (!dashes) +continue ; + + ring_name =realloc (ring_name, dashes - line); +strncpy (ring_name, line, dashes - line); + ring_name[dashes - line -1 ] ='\0' ; + + bb_matched =sscanf (dashes,"--- gtt_offset = 0x%08x\n" , + >t_offset); +if (bb_matched ==1 ) { +igt_assert (strstr (ring_name, expected_ring_name)); +igt_assert (gtt_offset == expected_offset); + +for (i =0 ; i <sizeof (batch) /4 ; i++) { +igt_assert (getline (&line, &line_size, file) >0 ); +snprintf (expected_line,sizeof (expected_line),"%08x : %08x" , +4 *i, batch[i]); +igt_assert (strstr (line, expected_line)); + } + bb_ok = true; +continue ; + } + + req_matched =sscanf (dashes,"--- %d requests\n" , + &requests); +if (req_matched ==1 ) { +igt_assert (strstr (ring_name, expected_ring_name)); +igt_assert (requests ==1 ); + +igt_assert (getline (&line, &line_size, file) >0 ); + items =sscanf (line," seqno 0x%08x, emitted %ld, tail 0x%08x\n" , + &seqno, &jiffies, &tail); +igt_assert (items ==3 ); + req_ok = true; +continue ; + } + + ringbuf_matched =sscanf (dashes,"--- ringbuffer = 0x%08x\n" , + >t_offset); +if (ringbuf_matched ==1 ) { +if (!strstr (ring_name, expected_ring_name)) +continue ; +igt_assert (req_ok); +
tests/gem_error_capture.c line 174
Quoted 10 lines
tests/gem_error_capture.c new file +10
-0+for (i =0 ; i < tail /4 ; i++) { +igt_assert (getline (&line, &line_size, file) >0 ); + items =sscanf (line,"%08x : %08x\n" , + &offset, &command); +igt_assert (items ==2 ); +if ((command &0x1F800000 ) ==MI_BATCH_BUFFER_START ) { +igt_assert (getline (&line, &line_size, file) >0 ); + items =sscanf (line,"%08x : %08x\n" , + &offset, &expected_addr); +igt_assert (items ==2 );
Should check for MAGIC_NUMBER here?
-Mika
Quoted 102 lines
tests/gem_error_capture.c new file +95
-0+ i++; + } + } +igt_assert (expected_addr == expected_offset); + ringbuf_ok = true; +continue ; + } + +if (bb_ok && req_ok && ringbuf_ok) +break ; + } +igt_assert (bb_ok && req_ok && ringbuf_ok); + +free (line); +free (ring_name); +close (debug_fd); +} + +static void test (int fd,uint32_t handle,unsigned ring_id,const char *ring_name) +{ +struct drm_i915_gem_execbuffer2 execbuf; +struct drm_i915_gem_exec_object2 exec; +uint64_t presumed_offset; + +gem_require_ring (fd, ring_id); + +clear_error_state (); + + exec.handle = handle; + exec.relocation_count =0 ; + exec.relocs_ptr =0 ; + exec.alignment =0 ; + exec.offset =0 ; + exec.flags =0 ; + exec.rsvd1 =0 ; + exec.rsvd2 =0 ; + + execbuf.buffers_ptr = (uintptr_t )&exec; + execbuf.buffer_count =1 ; + execbuf.batch_start_offset =0 ; + execbuf.batch_len =16 ; + execbuf.cliprects_ptr =0 ; + execbuf.num_cliprects =0 ; + execbuf.DR1 =0 ; + execbuf.DR4 =0 ; + execbuf.flags = ring_id; +i915_execbuffer2_set_context_id (execbuf,0 ); + execbuf.rsvd2 =0 ; + +gem_execbuf (fd, &execbuf); +gem_sync (fd, handle); + + presumed_offset = exec.offset; + +stop_rings (); + +gem_execbuf (fd, &execbuf); +gem_sync (fd, handle); + +igt_assert (rings_stopped () == false); +igt_assert (presumed_offset == exec.offset); + +check_error_state (ring_name, exec.offset); +} + +uint32_t handle; +int fd; + +igt_main +{ + igt_fixture { + fd =drm_open_any (); + + handle =gem_create (fd,4096 ); +gem_write (fd, handle,0 , batch,sizeof (batch)); + } + + igt_subtest("render") +test (fd, handle,I915_EXEC_RENDER ,"render ring" ); + + igt_subtest("bsd") +test (fd, handle,I915_EXEC_BSD ,"bsd ring" ); + + igt_subtest("blt") +test (fd, handle,I915_EXEC_BLT ,"blitter ring" ); + + igt_subtest("vebox") +test (fd, handle,I915_EXEC_VEBOX ,"video enhancement ring" ); + + igt_fixture { +gem_close (fd, handle); + +close (fd); + } +}Signature · 6 lines
1.9.0 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Hi Mika,
Quoted 19 lines
Quoted 6 lines
tests/gem_error_capture.c new file +6
-0+static bool rings_stopped (void ) +{ +int fd; +static char buf[128 ]; +unsigned long long val; +tests/gem_error_capture.c line 80
Quoted 10 lines
tests/gem_error_capture.c new file +10
-0+ fd =igt_debugfs_open ("i915_ring_stop" ,O_RDONLY ); +igt_assert (fd >=0 ); + +igt_assert (read (fd, buf,sizeof (buf)) >0 ); +close (fd); + +sscanf (buf,"%llx" , &val); + +return (bool )val; +}Please use
igt_set_stop_rings()andigt_get_stop_rings().
Quoted 1 line
Also consider stopping only the ring you are testing for.
Oops, I didn see these before. Ok, done in the new version.
Quoted 4 lines
Quoted 4 lines
tests/gem_error_capture.c new file +4
-0+for (i =0 ; i < tail /4 ; i++) { +igt_assert (getline (&line, &line_size, file) >0 ); + items =sscanf (line,"%08x : %08x\n" , + &offset, &command);
Quoted 8 lines
tests/gem_error_capture.c line 174
Quoted 6 lines
tests/gem_error_capture.c new file +6
-0+igt_assert (items ==2 ); +if ((command &0x1F800000 ) ==MI_BATCH_BUFFER_START ) { +igt_assert (getline (&line, &line_size, file) >0 ); + items =sscanf (line,"%08x : %08x\n" , + &offset, &expected_addr); +igt_assert (items ==2 );Should check for
MAGIC_NUMBERhere?
I'm doing it above already, inside bb_matched:
for (i = 0; i < sizeof(batch) / 4; i++) {
igt_assert(getline(&line, &line_size, file) > 0);
snprintf(expected_line, sizeof(expected_line), "%08x : %08x",
4*i, batch[i]);
igt_assert(strstr(line, expected_line));
}-- Oscar
PATCH 1/1 tests/gem_error_capture: Initial testcase for error state capture/dump
In reply to [email protected]From: Oscar Mateo <[email protected]>
Guarantees that error capture works at a very basic level.
v2: Also check that the ring object contains a reloc with MI_BB_START for the presumed batch object's address.
v3: Chris review comments:
- Move variables to local scope.
- Do not assume there is only one request.
- Some gen encode flags into the BB start address. Also, use igt_set/get_stop_rings as suggested by Mika Kuoppala.
--- tests/.gitignore | 1 + tests/Makefile.sources | 1 + tests/gem_error_capture.c | 254 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 256 insertions(+) create mode 100644 tests/gem_error_capture.c
tests/.gitignore +1 -0
@@ -27,6 +27,7 @@ gem_ctx_create gem_ctx_exec gem_double_irq_loop gem_dummy_reloc_loop +gem_error_capture gem_evict_alignment gem_evict_everything gem_exec_bad_domains
tests/Makefile.sources +1 -0
@@ -24,6 +24,7 @@ TESTS_progs_M = \ gem_ctx_bad_exec \ gem_ctx_exec \ gem_dummy_reloc_loop \ + gem_error_capture \ gem_evict_alignment \ gem_evict_everything \ gem_exec_bad_domains \
tests/gem_error_capture.c new file +254 -0
254 lines not on this page · raw message
Signature · 6 lines
1.9.0 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx