zstd: import usptream v1.5.2

Updates the kernel's zstd library to v1.5.2, the latest zstd release.
The upstream tag it is updated to is `v1.5.2-kernel`, which contains
several cherry-picked commits on top of the v1.5.2 release which are
required for the kernel update. I will create this tag once the PR is
ready to merge, until then reference the temporary upstream branch
`v1.5.2-kernel-cherrypicks`.

I plan to submit this patch as part of the v6.2 merge window.

I've done basic build testing & testing on x86-64, i386, and aarch64.
I'm merging these patches into my `zstd-next` branch, which is pulled
into `linux-next` for further testing.

I've benchmarked BtrFS with zstd compression on a x86-64 machine, and
saw these results. Decompression speed is a small win across the board.
The lower compression levels 1-4 see both compression speed and
compression ratio wins. The higher compression levels see a small
compression speed loss and about neutral ratio. I expect the lower
compression levels to be used much more heavily than the high
compression levels, so this should be a net win.

Level	CTime	DTime	Ratio
1	-2.95%	-1.1%	-0.7%
3	-3.5%	-1.2%	-0.5%
5	+3.7%	-1.0%	+0.0%
7	+3.2%	-0.9%	+0.0%
9	-4.3%	-0.8%	+0.1%

Signed-off-by: Nick Terrell <terrelln@fb.com>

1 files changed, 62 insertions, 16 deletions

diff --git a/lib/zstd/decompress/zstd_decompress.c b/lib/zstd/decompress/zstd_decompress.c
index 6928e85f9d19..b9b935a9f5c0 100644
--- a/lib/zstd/decompress/zstd_decompress.c
+++ b/lib/zstd/decompress/zstd_decompress.c
@@ -53,7 +53,6 @@
 *  Dependencies
 *********************************************************/
 #include "../common/zstd_deps.h"   /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */
-#include "../common/cpu.h"         /* bmi2 */
 #include "../common/mem.h"         /* low level memory routines */
 #define FSE_STATIC_LINKING_ONLY
 #include "../common/fse.h"
@@ -252,11 +251,11 @@ static void ZSTD_initDCtx_internal(ZSTD_DCtx* dctx)
     dctx->inBuffSize  = 0;
     dctx->outBuffSize = 0;
     dctx->streamStage = zdss_init;
-    dctx->legacyContext = NULL;
-    dctx->previousLegacyVersion = 0;
     dctx->noForwardProgress = 0;
     dctx->oversizedDuration = 0;
-    dctx->bmi2 = ZSTD_cpuid_bmi2(ZSTD_cpuid());
+#if DYNAMIC_BMI2
+    dctx->bmi2 = ZSTD_cpuSupportsBmi2();
+#endif
     dctx->ddictSet = NULL;
     ZSTD_DCtx_resetParameters(dctx);
 #ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
@@ -277,8 +276,7 @@ ZSTD_DCtx* ZSTD_initStaticDCtx(void *workspace, size_t workspaceSize)
     return dctx;
 }
 
-ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem)
-{
+static ZSTD_DCtx* ZSTD_createDCtx_internal(ZSTD_customMem customMem) {
     if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL;
 
     {   ZSTD_DCtx* const dctx = (ZSTD_DCtx*)ZSTD_customMalloc(sizeof(*dctx), customMem);
@@ -289,10 +287,15 @@ ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem)
     }
 }
 
+ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem)
+{
+    return ZSTD_createDCtx_internal(customMem);
+}
+
 ZSTD_DCtx* ZSTD_createDCtx(void)
 {
     DEBUGLOG(3, "ZSTD_createDCtx");
-    return ZSTD_createDCtx_advanced(ZSTD_defaultCMem);
+    return ZSTD_createDCtx_internal(ZSTD_defaultCMem);
 }
 
 static void ZSTD_clearDict(ZSTD_DCtx* dctx)
@@ -370,6 +373,19 @@ unsigned ZSTD_isFrame(const void* buffer, size_t size)
     return 0;
 }
 
+/*! ZSTD_isSkippableFrame() :
+ *  Tells if the content of `buffer` starts with a valid Frame Identifier for a skippable frame.
+ *  Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0.
+ */
+unsigned ZSTD_isSkippableFrame(const void* buffer, size_t size)
+{
+    if (size < ZSTD_FRAMEIDSIZE) return 0;
+    {   U32 const magic = MEM_readLE32(buffer);
+        if ((magic & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) return 1;
+    }
+    return 0;
+}
+
 /* ZSTD_frameHeaderSize_internal() :
  *  srcSize must be large enough to reach header size fields.
  *  note : only works for formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless.
@@ -497,7 +513,6 @@ size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t src
     return ZSTD_getFrameHeader_advanced(zfhPtr, src, srcSize, ZSTD_f_zstd1);
 }
 
-
 /* ZSTD_getFrameContentSize() :
  *  compatible with legacy mode
  * @return : decompressed size of the single frame pointed to be `src` if known, otherwise
@@ -532,6 +547,37 @@ static size_t readSkippableFrameSize(void const* src, size_t srcSize)
     }
 }
 
+/*! ZSTD_readSkippableFrame() :
+ * Retrieves a zstd skippable frame containing data given by src, and writes it to dst buffer.
+ *
+ * The parameter magicVariant will receive the magicVariant that was supplied when the frame was written,
+ * i.e. magicNumber - ZSTD_MAGIC_SKIPPABLE_START.  This can be NULL if the caller is not interested
+ * in the magicVariant.
+ *
+ * Returns an error if destination buffer is not large enough, or if the frame is not skippable.
+ *
+ * @return : number of bytes written or a ZSTD error.
+ */
+ZSTDLIB_API size_t ZSTD_readSkippableFrame(void* dst, size_t dstCapacity, unsigned* magicVariant,
+                                            const void* src, size_t srcSize)
+{
+    U32 const magicNumber = MEM_readLE32(src);
+    size_t skippableFrameSize = readSkippableFrameSize(src, srcSize);
+    size_t skippableContentSize = skippableFrameSize - ZSTD_SKIPPABLEHEADERSIZE;
+
+    /* check input validity */
+    RETURN_ERROR_IF(!ZSTD_isSkippableFrame(src, srcSize), frameParameter_unsupported, "");
+    RETURN_ERROR_IF(skippableFrameSize < ZSTD_SKIPPABLEHEADERSIZE || skippableFrameSize > srcSize, srcSize_wrong, "");
+    RETURN_ERROR_IF(skippableContentSize > dstCapacity, dstSize_tooSmall, "");
+
+    /* deliver payload */
+    if (skippableContentSize > 0  && dst != NULL)
+        ZSTD_memcpy(dst, (const BYTE *)src + ZSTD_SKIPPABLEHEADERSIZE, skippableContentSize);
+    if (magicVariant != NULL)
+        *magicVariant = magicNumber - ZSTD_MAGIC_SKIPPABLE_START;
+    return skippableContentSize;
+}
+
 /* ZSTD_findDecompressedSize() :
  *  compatible with legacy mode
  *  `srcSize` must be the exact length of some number of ZSTD compressed and/or
@@ -824,7 +870,7 @@ static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx,
         switch(blockProperties.blockType)
         {
         case bt_compressed:
-            decodedSize = ZSTD_decompressBlock_internal(dctx, op, (size_t)(oend-op), ip, cBlockSize, /* frame */ 1);
+            decodedSize = ZSTD_decompressBlock_internal(dctx, op, (size_t)(oend-op), ip, cBlockSize, /* frame */ 1, not_streaming);
             break;
         case bt_raw :
             decodedSize = ZSTD_copyRawBlock(op, (size_t)(oend-op), ip, cBlockSize);
@@ -976,7 +1022,7 @@ size_t ZSTD_decompress(void* dst, size_t dstCapacity, const void* src, size_t sr
 {
 #if defined(ZSTD_HEAPMODE) && (ZSTD_HEAPMODE>=1)
     size_t regenSize;
-    ZSTD_DCtx* const dctx = ZSTD_createDCtx();
+    ZSTD_DCtx* const dctx =  ZSTD_createDCtx_internal(ZSTD_defaultCMem);
     RETURN_ERROR_IF(dctx==NULL, memory_allocation, "NULL pointer!");
     regenSize = ZSTD_decompressDCtx(dctx, dst, dstCapacity, src, srcSize);
     ZSTD_freeDCtx(dctx);
@@ -1010,7 +1056,7 @@ static size_t ZSTD_nextSrcSizeToDecompressWithInputSize(ZSTD_DCtx* dctx, size_t
         return dctx->expected;
     if (dctx->bType != bt_raw)
         return dctx->expected;
-    return MIN(MAX(inputSize, 1), dctx->expected);
+    return BOUNDED(1, inputSize, dctx->expected);
 }
 
 ZSTD_nextInputType_e ZSTD_nextInputType(ZSTD_DCtx* dctx) {
@@ -1116,7 +1162,7 @@ size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, c
             {
             case bt_compressed:
                 DEBUGLOG(5, "ZSTD_decompressContinue: case bt_compressed");
-                rSize = ZSTD_decompressBlock_internal(dctx, dst, dstCapacity, src, srcSize, /* frame */ 1);
+                rSize = ZSTD_decompressBlock_internal(dctx, dst, dstCapacity, src, srcSize, /* frame */ 1, is_streaming);
                 dctx->expected = 0;  /* Streaming not supported */
                 break;
             case bt_raw :
@@ -1438,7 +1484,7 @@ size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx,
 ZSTD_DStream* ZSTD_createDStream(void)
 {
     DEBUGLOG(3, "ZSTD_createDStream");
-    return ZSTD_createDStream_advanced(ZSTD_defaultCMem);
+    return ZSTD_createDCtx_internal(ZSTD_defaultCMem);
 }
 
 ZSTD_DStream* ZSTD_initStaticDStream(void *workspace, size_t workspaceSize)
@@ -1448,7 +1494,7 @@ ZSTD_DStream* ZSTD_initStaticDStream(void *workspace, size_t workspaceSize)
 
 ZSTD_DStream* ZSTD_createDStream_advanced(ZSTD_customMem customMem)
 {
-    return ZSTD_createDCtx_advanced(customMem);
+    return ZSTD_createDCtx_internal(customMem);
 }
 
 size_t ZSTD_freeDStream(ZSTD_DStream* zds)
@@ -1708,7 +1754,8 @@ size_t ZSTD_sizeof_DStream(const ZSTD_DStream* dctx)
 size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize)
 {
     size_t const blockSize = (size_t) MIN(windowSize, ZSTD_BLOCKSIZE_MAX);
-    unsigned long long const neededRBSize = windowSize + blockSize + (WILDCOPY_OVERLENGTH * 2);
+    /* space is needed to store the litbuffer after the output of a given block without stomping the extDict of a previous run, as well as to cover both windows against wildcopy*/
+    unsigned long long const neededRBSize = windowSize + blockSize + ZSTD_BLOCKSIZE_MAX + (WILDCOPY_OVERLENGTH * 2);
     unsigned long long const neededSize = MIN(frameContentSize, neededRBSize);
     size_t const minRBSize = (size_t) neededSize;
     RETURN_ERROR_IF((unsigned long long)minRBSize != neededSize,
@@ -1842,7 +1889,6 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB
             DEBUGLOG(5, "stage zdss_init => transparent reset ");
             zds->streamStage = zdss_loadHeader;
             zds->lhSize = zds->inPos = zds->outStart = zds->outEnd = 0;
-            zds->legacyVersion = 0;
             zds->hostageByte = 0;
             zds->expectedOutBuffer = *output;
             ZSTD_FALLTHROUGH;