From 179f7afb9b0782d2f89367de81b7a0bb26b87e5b Mon Sep 17 00:00:00 2001 From: Nigel Tao Date: Sat, 22 Dec 2018 14:45:42 +1100 Subject: [PATCH] Have lzw rewind in and flush out on a bad code MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit name old speed new speed delta wuffs_lzw_decode_20k/clang 214MB/s ± 0% 204MB/s ± 2% -4.62% (p=0.008 n=5+5) wuffs_lzw_decode_100k/clang 369MB/s ± 0% 337MB/s ± 2% -8.72% (p=0.008 n=5+5) wuffs_lzw_decode_20k/gcc 232MB/s ± 1% 236MB/s ± 0% +1.99% (p=0.008 n=5+5) wuffs_lzw_decode_100k/gcc 384MB/s ± 1% 396MB/s ± 1% +3.00% (p=0.008 n=5+5) wuffs_gif_decode_1k_bw/clang 236MB/s ± 0% 230MB/s ± 0% -2.67% (p=0.008 n=5+5) wuffs_gif_decode_1k_color/clang 118MB/s ± 1% 120MB/s ± 0% +1.10% (p=0.008 n=5+5) wuffs_gif_decode_10k_bgra/clang 562MB/s ± 0% 562MB/s ± 0% ~ (p=0.690 n=5+5) wuffs_gif_decode_10k_indexed/clang 149MB/s ± 0% 149MB/s ± 0% ~ (p=0.421 n=5+5) wuffs_gif_decode_100k_artificial/clang 487MB/s ± 0% 459MB/s ± 1% -5.63% (p=0.008 n=5+5) wuffs_gif_decode_100k_realistic/clang 183MB/s ± 0% 174MB/s ± 0% -4.52% (p=0.008 n=5+5) wuffs_gif_decode_1000k/clang 185MB/s ± 0% 177MB/s ± 1% -4.61% (p=0.008 n=5+5) wuffs_gif_decode_anim_screencap/clang 1.02GB/s ± 0% 0.97GB/s ± 0% -5.30% (p=0.008 n=5+5) wuffs_gif_decode_1k_bw/gcc 250MB/s ± 0% 249MB/s ± 0% -0.63% (p=0.016 n=5+5) wuffs_gif_decode_1k_color/gcc 133MB/s ± 0% 133MB/s ± 1% ~ (p=0.151 n=5+5) wuffs_gif_decode_10k_bgra/gcc 543MB/s ± 1% 543MB/s ± 0% ~ (p=0.548 n=5+5) wuffs_gif_decode_10k_indexed/gcc 167MB/s ± 1% 166MB/s ± 1% -0.83% (p=0.032 n=5+5) wuffs_gif_decode_100k_artificial/gcc 465MB/s ± 1% 465MB/s ± 1% ~ (p=0.841 n=5+5) wuffs_gif_decode_100k_realistic/gcc 190MB/s ± 0% 190MB/s ± 1% ~ (p=0.222 n=5+5) wuffs_gif_decode_1000k/gcc 192MB/s ± 1% 193MB/s ± 0% ~ (p=0.421 n=5+5) wuffs_gif_decode_anim_screencap/gcc 1.00GB/s ± 1% 1.00GB/s ± 1% ~ (p=0.841 n=5+5) --- release/c/wuffs-unsupported-snapshot.h | 14 ++++++-- std/lzw/decode_lzw.wuffs | 9 ++++- test/c/std/lzw.c | 49 +++++++++++++++++++++++++- 3 files changed, 68 insertions(+), 4 deletions(-) diff --git a/release/c/wuffs-unsupported-snapshot.h b/release/c/wuffs-unsupported-snapshot.h index e1907eb6..58b7e1ca 100644 --- a/release/c/wuffs-unsupported-snapshot.h +++ b/release/c/wuffs-unsupported-snapshot.h @@ -2628,6 +2628,7 @@ struct wuffs_lzw__decoder__struct { uint32_t v_save_code; uint32_t v_prev_code; uint32_t v_width; + bool v_bad; uint32_t v_j; uint32_t v_bits; uint32_t v_n_bits; @@ -7145,6 +7146,7 @@ wuffs_lzw__decoder__decode(wuffs_lzw__decoder* self, uint32_t v_save_code; uint32_t v_prev_code; uint32_t v_width; + bool v_bad; uint32_t v_j; uint32_t v_bits; uint32_t v_n_bits; @@ -7181,6 +7183,7 @@ wuffs_lzw__decoder__decode(wuffs_lzw__decoder* self, v_save_code = self->private_impl.c_decode[0].v_save_code; v_prev_code = self->private_impl.c_decode[0].v_prev_code; v_width = self->private_impl.c_decode[0].v_width; + v_bad = self->private_impl.c_decode[0].v_bad; v_j = self->private_impl.c_decode[0].v_j; v_bits = self->private_impl.c_decode[0].v_bits; v_n_bits = self->private_impl.c_decode[0].v_n_bits; @@ -7192,6 +7195,7 @@ wuffs_lzw__decoder__decode(wuffs_lzw__decoder* self, v_lm1_b = 0; v_lm1_a = 0; } else { + v_bad = false; } switch (coro_susp_point) { WUFFS_BASE__COROUTINE_SUSPENSION_POINT_0; @@ -7205,6 +7209,7 @@ wuffs_lzw__decoder__decode(wuffs_lzw__decoder* self, v_save_code = v_end_code; v_prev_code = v_end_code; v_width = (v_literal_width + 1); + v_bad = 0; v_j = 0; while (v_j < v_clear_code) { self->private_impl.f_lm1s[v_j] = 0; @@ -7334,8 +7339,8 @@ wuffs_lzw__decoder__decode(wuffs_lzw__decoder* self, v_prev_code = v_code; } } else { - status = wuffs_lzw__error__bad_code; - goto exit; + v_bad = true; + goto label_1_break; } if (v_j > 4095) { while (v_n_bits >= 8) { @@ -7375,6 +7380,10 @@ wuffs_lzw__decoder__decode(wuffs_lzw__decoder* self, goto suspend; } } + if (v_bad) { + status = wuffs_lzw__error__bad_code; + goto exit; + } goto ok; ok: @@ -7391,6 +7400,7 @@ suspend: self->private_impl.c_decode[0].v_save_code = v_save_code; self->private_impl.c_decode[0].v_prev_code = v_prev_code; self->private_impl.c_decode[0].v_width = v_width; + self->private_impl.c_decode[0].v_bad = v_bad; self->private_impl.c_decode[0].v_j = v_j; self->private_impl.c_decode[0].v_bits = v_bits; self->private_impl.c_decode[0].v_n_bits = v_n_bits; diff --git a/std/lzw/decode_lzw.wuffs b/std/lzw/decode_lzw.wuffs index 3ff7e5a5..afd11364 100644 --- a/std/lzw/decode_lzw.wuffs +++ b/std/lzw/decode_lzw.wuffs @@ -49,6 +49,8 @@ pub func decoder.decode!??(dst base.io_writer, src base.io_reader) { var prev_code base.u32[..4095] = end_code var width base.u32[..12] = literal_width + 1 + var bad base.bool + // First, use the j variable to initialize some state. var j base.u32[..8191] while j < clear_code { @@ -208,7 +210,8 @@ pub func decoder.decode!??(dst base.io_writer, src base.io_reader) { } } else { - return "?bad code" + bad = true + break } // Flush the output if it could be too full to contain the entire @@ -247,6 +250,10 @@ pub func decoder.decode!??(dst base.io_writer, src base.io_reader) { this.flush_j = j this.auto_flush!??(dst:args.dst) } + + if bad { + return "?bad code" + } } pri func decoder.auto_flush!??(dst base.io_writer) { diff --git a/test/c/std/lzw.c b/test/c/std/lzw.c index cdcf3b9d..21338b6c 100644 --- a/test/c/std/lzw.c +++ b/test/c/std/lzw.c @@ -197,6 +197,52 @@ const char* test_wuffs_lzw_decode_pi() { "test/data/pi.txt", 100003, 0, 0); } +const char* test_wuffs_lzw_decode_output_bad() { + CHECK_FOCUS(__func__); + + wuffs_base__io_buffer got = ((wuffs_base__io_buffer){ + .data = global_got_slice, + }); + wuffs_base__io_buffer src = ((wuffs_base__io_buffer){ + .data = global_src_slice, + }); + + // Set up src to be 20 bytes long, starting with three 8-bit literal codes + // (0x41, 0x42, 0x43) then a bad 8-bit code 0xFF. Decoding should produce 3 + // bytes and consume 4 bytes. + src.meta.wi = 20; + int i; + for (i = 0; i < src.meta.wi; i++) { + src.data.ptr[i] = 0; + } + src.data.ptr[0] = 0x41; + src.data.ptr[1] = 0x42; + src.data.ptr[2] = 0x43; + src.data.ptr[3] = 0xFF; + + wuffs_lzw__decoder dec = ((wuffs_lzw__decoder){}); + const char* z = + wuffs_lzw__decoder__check_wuffs_version(&dec, sizeof dec, WUFFS_VERSION); + if (z) { + RETURN_FAIL("check_wuffs_version: \"%s\"", z); + } + wuffs_lzw__decoder__set_literal_width(&dec, 7); + + z = wuffs_lzw__decoder__decode(&dec, wuffs_base__io_buffer__writer(&got), + wuffs_base__io_buffer__reader(&src)); + if (z != wuffs_lzw__error__bad_code) { + RETURN_FAIL("decode: \"%s\"", z); + } + + if (got.meta.wi != 3) { + RETURN_FAIL("got.meta.wi: got %d, want 3", (int)(got.meta.wi)); + } + if (src.meta.ri != 4) { + RETURN_FAIL("src.meta.ri: got %d, want 4", (int)(src.meta.ri)); + } + return NULL; +} + const char* test_wuffs_lzw_decode_output_empty() { CHECK_FOCUS(__func__); @@ -208,7 +254,7 @@ const char* test_wuffs_lzw_decode_output_empty() { }); // Set up src to be 20 bytes long, starting with the 9-bit end code 0x101. - // Decoding should consume only 2 of those bytes. + // Decoding should produce 0 bytes and consume 2 bytes. src.meta.wi = 20; int i; for (i = 0; i < src.meta.wi; i++) { @@ -308,6 +354,7 @@ proc tests[] = { test_wuffs_lzw_decode_bricks_nodither, // test_wuffs_lzw_decode_many_big_reads, // test_wuffs_lzw_decode_many_small_writes_reads, // + test_wuffs_lzw_decode_output_bad, // test_wuffs_lzw_decode_output_empty, // test_wuffs_lzw_decode_pi, //